Alternative for the Cox regression model: using parametric models to analyze the survival of cancer patients

Although the Cox proportional hazard regression is the most popular model for analyzing the prognostic factors on survival of cancer patients, under certain circumstances, parametric models estimate the parameter more efficiently than the Cox model. The aim of this study was to compare the Cox regression model with parametric models in patients with gastric cancer who registered at Taleghani hospital, Tehran, Iran. In a retrospective cohort study, 746 patients with gastric cancer were studied from February 2003 through January 2007. Gender, age at diagnosis, distant metastasis, extent of wall penetration, tumor size, histology type, tumor grade, lymph node metastasis and pathologic stage were selected as prognosis, and entered to the models. Lognormal, Exponential, Gompertz, Weibull, Loglogistic and Gamma regression were performed as parametric models, and Akaike Information Criterion [AIC] were used to compare the efficiency of the models. Based on AIC, Log logistic is an efficient model. Log logistic analysis indicated that wall penetration and presence of pathologic distant metastasis were potential risks for death in full and final model analyses. In the multivariate analysis, all the parametric models fit better than Cox with respect to AIC; and the log logistic regression was the best model among them. Therefore, when the proportional hazard assumption does not hold, these models could be used as an alternative and could lead to acceptable conclusions

Family history as a risk for upper gastrointestinal tract cancer: a case control study

Although, family history of cancer is an important risk factor for upper gastrointestinal cancers development, but limited information is available on the upper gastrointestinal cancers associated with family history in Iran. The purpose of this study was to define upper gastrointestinal cancers risk associated with family history of cancer. This study was conducted as a case control study. A total number of 1,010 cases of upper gastrointestinal cancer and 1,010 healthy controls were recruited. For family history of cancer, questions were asked about any malignant tumor in first and second degree relatives. Adjusted odds ratio estimates for the association family history and upper gastrointestinal cancers risk and corresponding 95% confidence intervals were obtained. A family history of any malignant tumor in relatives was associated with 1.3 fold increased risks of upper gastrointestinal cancers. A first-degree family history of esophageal and gastric cancer was significantly associated with upper gastrointestinal cancers development, with an adjusted OR of 4.7 [CI 95%: 2.6-8.4]. Our findings suggested that risk for upper gastrointestinal cancers increases among individuals with family history of cancer. Therefore, appropriate screening strategies especially in relatives of patients should be considered to prevent and control of disease

Epidemiological differences between colon cancer and rectum cancer

Clinical and epidemiological variation was seen between the colon cancer [CC] and rectum cancer [RC]. So, there is not so much data available about the epidemiological and clinicopathological differences and prognostic factors regarding to CC and CR in Iran, we aimed to perform this study. All cases of CC and RC referred to oncology and gastroenterology wards of Taleghani General Hospital, Teheran, Iran between 2002 and 2008 were retrospectively reviewed. The research group were reviewed all medical records in the study period for collecting the required data. All patients under study were followed up until end day of 2008 [closed day] from their diagnosis. There are 856 cases of CC and 427 cases of RC. Mean survival time of CC cases was relatively higher than RC cases [P <0.05]. Regarding to the age at diagnosis, about 42% of CC and 42.6% of RC patients was diagnosed less than 50 years of age. Positive family history of any cancer was relatively higher in CC [40.0%] patients than RC [31.0%] patients [P<0.05]. significant difference was seen between CC and RC regarding to depth of tumor invasion, pathologic stage and type of first treatment. RC patient were diagnosed in more advanced pathologic stages. Regarding to histology type of tumor 75.0% of CC cases and 79.4% of RC cases was adenocarcinoma. Abdominal pain [74.4%] and blood per rectum [89.7%] were the most prevalent symptoms mentioned by patients for CC and RC, respectively. Distant metastasis, lymph node metastasis, lower BMI and poor grading of tumor was related to increased risk of death due to CC. Regarding to RC, only pathologic stage was determine as prognostic factor. Results of this study emphasis that RC has a poorer prognosis comparing to CC. Up to 42 percent of patients with CC and RC are lower than 50 years of age. Patterns of CC versus RC indicate major variations in demographic and clinicopathologic characteristics that suggest possible differences in etiology and pathogenesis. So we suggest that for the analysis of cancer data, CC and RC should be investigated as separate cancers and not to be as colorectal cancer. Abdominal pain and blood per rectum should be emphasis for detection of CC and RC, respectively

Prevalence of positive family history of colorectal cancer in the Iranian general population

Individuals with a positive family history of colorectal cancer have an increased risk of developing this type of cancer. The number of affected relatives and the age at diagnosis are two factors that increase the risk of colorectal cancer. The aim of this study was to assess the prevalence of a positive family history of colorectal cancer in a random sample among the Iranian general population. Five thousand five hundred [5500] subjects' aged>/-20 years were randomly selected by cluster sampling and invited to participate in an interview about the occurrence of colorectal cancer in their first- or second-degree relatives. Of all the responders, 162 [2.9%] subjects reported a positive family history of colorectal cancer; 71 [1.24%] reported having one first-degree relative with colorectal cancer diagnosed before the age of 50; or reported two or more first-degree relatives with colorectal cancer. In addition, 83 [1.51%] and 14 [0.25%] subjects reported having one and two or more second-degree relatives with colorectal cancer respectively. The prevalence of a positive family history of colorectal cancer in Iran is lower than the United States and European countries. Identifying high-risk population for colorectal cancer and encouraging them to participate in surveillance protocols is the first step in targeting preventive measures

Positive association between diabetes mellitus and risk of colorectal cancer

In recent years, many studies employed and found an association between type 2 diabetes mellitus and colorectal cancer. Although increased risk of colorectal cancer in individuals with Non Insulin Dependent Diabetes Mellitus [NIDDM] has been observed in previous studies, limited information is available on the colorectal cancer associated with NIDDM in Iran. The purpose of this study was to define colorectal cancer risk associated with diabetes mellitus. The present study was designed as an unmatched case control study. Cases were 393 patients with histologically confirmed colorectal carcinomas and 393 controls were randomly selected among the healthy participants in a health survey. To control potential confounding factors such as sex, age, smoking habits and Body Mass Index [BMI], multiple logistic regression model was fitted to obtain Odds Ratio of colorectal cancer and the corresponding 95% CIs, according to history of diabetes mellitus. Overall, 86 [10%] cases versus 15 [1.7%] controls gave a history of diabetes mellitus. The corresponding multivariate OR was 6.77 [CI 95%: 3.84- 11.92] indicating that having a positive history of diabetes mellitus increases one's risk of colorectal cancer about six-fold. The risk of colorectal cancer was slightly increased for women [p<0.05]. Current smokers were at a higher CRC risk [OR=2.83, CI95%:2.13-3.76] than never smokers. We found a strong positive association between NIDDM and prevalent colorectal cancer. In summary, these findings provide further indirect epidemiological evidence for the hypothesis that hyperinsulinaemia may be important in the development of colorectal cancer

Clinicopathological features of gastric cancer: a study based on cancer registry data

To analysis the epidemiological and clinopathological aspects of gastric cancer. We retrospectively studied the characteristics of 752 gastric carcinoma patients registered in cancer registry center of Taleghani hospital from 2001 to 2006. All the patients confirmed gastric cancer histopathologically. Univariate methods [Mann-Whitney U-test, and chi-square] were used for analysis. Among 752 cases, 535 [71.1%] were male. Sex ratio [male: female] was 2.49:1. The mean age of the patients was 59.7 years [SD = 12.9] at the time of the diagnosis. Weight loss, as a frequent symptom at the time of diagnosis, was observed in 57.7% of the cases. Most of the patients were diagnosed with advanced pathologic stage. In addition, tumor grading was poorly differentiated in most cases [28.7%]; and in 49.9% of the patients, the tumors were located in the stomach, NOS; and distant metastasis was observed in 24.6% of the cases. Based on the results, earlier detection in younger ages and in primary stages of tumor, and subsequent higher quality care to cure GC and increase patients' life expectancy are recommended

Pathology and prognosis of colorectal cancer

Colorectal cancer [CRC] is one of the most common cancers in the world. During the past decades, survival of colorectal cancer patients has improved worldwide, however, it is not clear what factors have contributed to this development. This study was designed to evaluate the prognostic impact of a wide spectrum of pathologic parameters on survival rate in patients with colorectal cancer. 1127 patients with colorectal cancer who registered in one cancer registry in Iran were followed from their diagnostic date to Jan 1, 2007 [as failure time]. Overall survival time was calculated by Kaplan-Meier method. The Cox proportional hazard model was used to identify the pathologic factors that could independently influence survival. The overall survival rate at 5 years after diagnosis was 61%. Histology grade, status of regional lymph node metastasis, distant metastasis and pathologic tumor stage were related to survival rate according to univariate analysis. Nevertheless, in multivariate analysis, only histology grade, distant metastasis and tumor size had influence on survival of colorectal cancer patients. Generally the prognosis of disease is not poor; however, distant metastasis, poor differentiation and higher tumor size should be considered to have additional risks of death in colorectal cancer

Estimation of survival and factors related to it in Esophageal Cancer using actuarial life-table mathod

Esophageal cancer is the 6th common cancer in Iran. Affected patients have a relatively short lifetime, as this cancer is usually diagnosed in advanced stages. This study was conducted to estimate survival and factors related to it in patients with esophageal cancer. All patients definitely diagnosed as esophageal cancer, registered in Fars Cancer Registry Center, were investigated. Required information was gathered from cancer registry forms and other resources. 1- to 5-year survival rates were estimated using life-table method. Wilcoxon test was used to compare survival rates between subgroups. All data were analyzed by SPSS software. Of 161cases, 61.5% were male. Mean age of patients was 64.6 +/- 13.10 years and 57.6 +/- 11.9 years in male and female, respectively. Overall survival rates at 1- to 5-years after diagnosis were: 58, 33, 18, 15 and 9 percent, respectively. Factors such as age at time of diagnosis, Presence of metastasis, histologic grade, type of first treatment, occupation in males, ethnicity and number of household showed to have a significant effect on prognosis. According to this study and similar studies carried out, screening and early detection of patients [in lower ages, and lower stages] is emphasized

[Reliability and validity of the QLQ-C30 questionnaire in cancer patients]

Recently, it has been recognized that a more comprehensive assessment of the cancer patient is necessary and that the evaluation of outcomes must move beyond traditional biomedical endpoints to include assessments of the impact of disease and its treatment on patients' quality of life. European Organization for Research and Treatment of Cancer has developed a 30-item quality of life questionnaire to obtain information about the impact of disease and treatment on the daily living of cancer patients. This questionnaire was translated into many languages and used in various countries. But, validity and reliability of version 3.0 has not yet been checked for use with Iranian patients. The aims of the present study are to evaluate the reliability and validity of the QLQ-C30 [version 3.0] questionnaire. We conducted an analytical cross-sectional study on 132 random samples of breast cancer patients for determination of validity and reliability of QLQ-C30 [version 3.0] questionnaire. The questionnaire was completed for all patients that referred to chemotherapy ward in Namazi Hospital, Shiraz, during 19 Dec 2005 -19 Feb 2006. Reliability was evaluated through the internal consistency of multi-item subscales. Pearson's correlations of an item with its own scale [corrected for overlap] and other scales were calculated to evaluate convergent and discriminate validity. Clinical validity was evaluated by known-group comparisons using ANOVA and Kruskal Wallis tests. Data were then analyzed with SPSS software. In the reliability analysis, most scales fitted the criteria [Alpha>0.70] except the fatigue, pain, nausea and vomiting scales. Convergent validity was evidenced by item own subscale correlation above 0.40 for all multi-item subscales. Item discriminate validity was successful in all analyses except for item 4 of the physical functioning scale. Results of the known group based analysis show significant differences in QLQ-C30 functioning and symptom scores, where patients with higher grade have the worst [p<0.05]. Iranian version of EORTC QLQC30 is a reliable and valid QOL measure for cancer patients which indicate that it can be used in clinical and epidemiological cancer studies

[Quality of life and related factors in breast cancer patients under chemotherapy]

Today, the quality of life study has an important role in health care especially in chronic diseases, in clinical judgment and in medical resources supplying. Malignancies have a clinical and health importance in the world and Iran. Breast cancer has first order among women's malignancies. Now, survival rate for this cancer is long. However Breast cancer has several complications that affected patient's life. It is necessary that we studying the quality of life and related factors among this patients. We conducted a cross-sectional study on 119 breast cancer patients that admitted and treated in chemotherapy ward of Namazi hospital in Shiraz.We used QLQ-C30 and QLQ-BR23 to assessment quality of life in these patients. We used univariate nonparametric tests and multiple linear regression model to identify associations between dependent variables and the quality of life and it's different scales.All calculation performed by using SPSS.V.13. Mean age of patients was 48.27 +/- 11.42 with quality of life total score 64.92 +/- 24.28. Univariate analysis showed that occupation,duration of disease,grade of tumor, physical, emotional and cognitive functioning, also, symptoms such as fatigue,pain, nausea and vomiting, insomnia, constipation and financial difficulties, perspective were associated to quality of life. [P< 0.05]. But in multivariate analysis, only occupation, menopause status, dyspnea, grade of tumor and financial difficulties perspective were related to quality of life [p<0.05]. With this finding, It is recommended that financial supports, also early detection are necessary for improvement of quality of life in these patients