Descriptive Epidemiology of Colorectal Cancer in University Malaya Medical Centre , 2001 to 2010

Colorectal cancer is a major public health problem globally. Overall, it ranks as the third most frequent cancer worldwide, and the third and second most frequent cancer in men and women respectively (Jemal et al., 2011). The incidence of colorectal cancer in Asia, and particularly South East Asia, has assumed the global trend. Presently, colorectal cancer is the third most frequent cancer in Asian men and women (Pourhoseingholi, 2012). Colorectal cancer is the second leading cancer among the general population in Malaysia, according to the Ministry of Health annual report 2011 Yahaya et al (2010). Colorectal cancer accounts for 12.7% of all cancers diagnosed, comprising 15.7% and 10.4% of male and female cancers respectively. Nearly equal numbers of colon and rectal cancers were diagnosed (4,547; 15.7 per 100,000 versus 4,689; 16.2 per 100,000) with similar mortality rates (241; 0.83 per 100,000 versus 229; 0.79 per 100,000).


Introduction
Colorectal cancer is a major public health problem globally.Overall, it ranks as the third most frequent cancer worldwide, and the third and second most frequent cancer in men and women respectively (Jemal et al., 2011).The incidence of colorectal cancer in Asia, and particularly South East Asia, has assumed the global trend.Presently, colorectal cancer is the third most frequent cancer in Asian men and women (Pourhoseingholi, 2012).
Colorectal cancer is the second leading cancer among the general population in Malaysia, according to the Ministry of Health annual report 2011Yahaya et al (2010)).Colorectal cancer accounts for 12.7% of all cancers diagnosed, comprising 15.7% and 10.4% of male and female cancers respectively.Nearly equal numbers of colon and rectal cancers were diagnosed (4,547; 15.7 per 100,000 versus 4,689; 16.2 per 100,000) with similar mortality rates (241; 0.83 per 100,000 versus 229; 0.79 per 100,000).

Descriptive Epidemiology of Colorectal Cancer in University Malaya Medical Centre, 2001 to 2010
Bello Arkilla Magaji 1,2,3 *, Foong Ming Moy 1 , April Camilla Roslani 2,3 , Chee Wei Law 2,3 The upward trend in incidence of colorectal cancers in Asia is influenced by lifestyle and environmental factors such as smoking, obesity, physical inactivity and, to some extent, the ethnic background of the patients (Chong et al., 2009;Tsukuma et al., 2011).This suggests an urgent need to understand the differential burden of colorectal cancer in countries in Asia (Hyodo et al., 2010).Therefore, study of the epidemiologic characteristics of colorectal cancer in Malaysia is both necessary and timely, in order to guide the establishment of screening services, plan colorectal cancer prevention public health education and awareness campaigns and improve treatment outcomes.
Consequently, we aimed to study the epidemiologic characteristics, survival and health-related quality of life (HRQOL) of the colorectal cancer patients treated at the University Malaya Medical Centre (UMMC) over a tenyear period.This paper reports the first components of the research whose aim was to determine the demographic, clinical and treatment characteristics of colorectal cancer patients treated at the UMMC between January 2001 and  (Wendy and Radzi, 2008).Each patient's unique national identity card as well as hospital registration number was used to identify and link their medical records.In this study, only Malaysian citizens or permanent residents with a diagnosis of colorectal cancer were included.

Materials and Methods
Demographic variables studied were: age at the time of diagnosis, gender and ethnicity.The clinical characteristics included are; tumor site, Dukes stage, and tumor cellular differentiations.The treatment variables were surgery, type of surgery, urgency of surgery, chemotherapy and radiotherapy.
Exploratory data analysis was done to identify the missing data and guide the choice of statistical test.Age and other quantitative variables were summarized using mean and standard deviations, while gender, ethnicity and other categorical variables were summarized using proportions and percentages.The Chi squared (Chi2) test was used to compare differences in proportions between categorical variables.Differences in means for two or more numeric variables were examined using t-test and ANOVA respectively.Where differences were observed, post hoc (Bonferroni) analysis was performed to understand the nature of the differences.A two-tailed significance level of 0.05 was used.Analysis was done using SPSS version 21.0 for windows, (SPSS Inc., Chicago, Illinois, USA).This project was approved by the ethics committee of the University Malaya Medical Centre (PPUM/UPP/300/02/02, MEC 770.2).

Demographic characteristics of the patients
The final sample consisted of 1212 colorectal cancer patients.The male to female ratio was 1.2:1.Sixty-seven per cent were Chinese, 18%, were Malays, 13% Indians, and 2%other ethnic groups.
At the time of diagnosis, 26% had hypertension and 16% had diabetes mellitus.Pre-operative carcinoembryonic antigen was elevated in 21% of patients.Highgrade tumors (i.e.poorly and undifferentiated tumors) were present in 5% of patients.

Treatment characteristics
Treatment characteristics are shown in Table 4. Surgery (82%) was the most common treatment strategy used during the study period.Of the surgeries performed, 11% were emergency operation.The commonest procedures were anterior resection (all types; 26%); followed by right hemi-colectomy (all types; 22%); and abdomino-perineal resection (10%).Further analyses showed that 40.3% had surgery alone, 32.4% had surgery and chemotherapy or radiotherapy, 8.3% had chemotherapy and radiotherapy, 9.6% had surgery, chemotherapy and radiotherapy and 9.4% had palliative care only.

Discussion
Colorectal cancer is no longer predominantly a disease of the Western world.Colon and rectal cancers are now a major public health problem in Malaysia, yet there is little data on its management, as national data is currently focused on epidemiology, and is limited by underreporting.Hospital-based clinical and epidemiologic    1, the ratio of high grade tumors between the younger age group (≤39 years) and the older (≥ 40 years) was 2.5:1 (15.8% vs 6.3%); this difference was statistically significant (χ 2 =7.35,P< 0.007).Males (57%) had low grade tumors compared to female (44%).Conversely, more females (58%) had high grade tumors than male (42%).The difference was research remains a valuable means of understanding the current burden of colorectal cancers in countries with limited data at the national level.In addition to being the primary sources for national level epidemiological data, hospital-based data provides more clinically-rich information on patients, including treatment and outcomes of treatment.It therefore contributes significantly in the resource management for patients, hospitals and the country (Nadathur, 2010).In this research, we studied a ten-year cohort of colorectal cancer patients treated at the UMMC.Findings were compared with previous literature from Malaysia and other countries.
Regarding the demographic, clinical and treatment characteristics, our patients are generally similar to other patients in Malaysia and other South East Asian countries.We found mean age to be around 60 years, more male patients and over two-thirds are of Chinese ethnicity.These findings are similar to the earlier studies in Malaysia (Zarihah et al., 2003;Goh et al., 2005;Ooi et al., 2005;Chong et al., 2009;Hassan et al., 2010).
The mean age was slightly lower than reported in Singapore, Japan and the Western world (Vines et al., 2003).Similar to earlier reports, this study reiterates that a significant proportion of Malays present at younger age compared to the Chinese and Indians (King and Kutty, 1971).A recent study from Jeddah Saudi Arabia reports mean age of 58 for male and 53 for female (Mosli and Al-Ahwal, 2012) and Iran mean age of 57.7 for male and 56.6 for female (Hajmanoochehri et al., 2014).Possible explanations are yet to be unveiled.However, genetic factors might be responsible.These findings are important when screening services are considered in the Malaysian multi-ethnic society.Future research is needed to understand the risk factors and pathogenesis of colorectal cancer among the different ethnic groups in Malaysia.
No significant difference between males and females was observed in our study.This is similar to a recent study from Iran (Hajmanoochehri et al., 2014).Previous studies, however, suggested that males have a higher risk for colorectal cancer (Mosli and Al-Ahwal, 2012;Shah et al., 2014).This is said to be due to exposure to dietary, lifestyle risk sex-specific exposure to/or metabolism of environmental risk factors, differences in screening experiences and access to medical care (de Kok et al., 2008;Koo and Leong, 2010).It is possible that males in our population are less likely to demonstrate health-related help-seeking behavior, and thus are under-diagnosed compared with females.
The higher incidence in Chinese patients is in line with the earlier observation that ethnicity may play a role in the etiology of colorectal cancers in Asia (Yee et al., 2009;Pourhoseingholi, 2012).Although previous Malaysian national cancer registry reports have shown the Chinese having the highest incidence of colorectal cancer (Zarihah et al., 2003;Lim et al., 2004;Zainal Ariffin et al., 2011;Shah et al., 2014), local registries have sometimes contradicted this; this was attributed to dissimilarities in the local ethnic distribution.However, given the proportion of Chinese colorectal cancers far exceeds their proportion in our catchment area, it is likely to reflect a true increased risk.Similar findings were reported from Brunei (Chong et al., 2009).As expected, Indians have the lowest figures.Pathy et al. (2012) concluded that colorectal cancer incidence among Indians remains the lowest compared to other races in Asia.
The clinical features observed in our cohort of patients reflect the current trend in colorectal cancer in most developing nations.As expected, patients in UMMC had more rectal cancers (involving recto-sigmoid, rectum and ano-rectal sites).These cancers are as well more among the male population compared to female.Also, rectal cancers were more in the Malay ethnic groups.These findings are similar to the findings of a recent study from a review of surgeries relating to colorectal cancers in the northern state of Kedah (Hassan et al., 2011).
One of the most striking observations is that twothirds of patients had advanced stage (Dukes' C and D), irrespective of their gender, age or ethnicity.Law et al (2009) reported that the ethnic background of rectal cancer patients in UMMC was not associated with prehospital delays, and others have suggested that access to health care, rather than ethnicity or racial background, determines stage at presentation (Hassan et al., 2009;Laiyemo et al., 2010).Similarly, African-Americans are at increased risk of presenting with late stage cancers, and poorer overall outcomes, compared to Caucasians, mainly due to difficulties with health care access (Sabounchi et al., 2012).Other factors affecting delay in diagnosis of colorectal cancer in our settings are poor knowledge of the symptoms, attitudes to screening as well as health care personnel failure to suspect colorectal cancer in patients presenting with a rectal bleeding (Hashim et al., 2010;Kong et al., 2010;Hashim et al., 2011).Advanced colorectal cancer disease in 2/3rd of the patients was also reported among patients in other Asian counties (Amin et al., 2012;Hajmanoochehri et al., 2014) Our younger (≤ 40 years) patients had significantly higher tumor grade compared to older age groups.This is in line with previous studies where colorectal cancers were found to be more aggressive, spread faster and are associated with a poor prognosis in younger age groups (Chou et al., 2011;Amin et al., 2012).The higher proportion of young Malays affected suggests that this section of the population may need to be screened at a younger age than the Asian Pacific guidelines suggest.
Treatment delivered in our center is in accordance with evidence-based recommendations.The use of surgery as the primary treatment modality, and indications for adjuvant chemo-and radiotherapy, are well established in the literature.As summarized by the National Institute of health and Care Excellence full guidelines 2011, the choice of treatment modality depends on presentation as emergency or otherwise, pre-treatment staging, presence of co-morbidities, metastasis, tumor site, level of risk for local recurrences (rectal cancer).The decision on the final treatment modality will depends on proper assessment of these parameters and decision of both the management team and a well-informed patient.
Not surprisingly, the majority of our patients received surgery, even those with metastatic disease, where the aim was either curative resection or surgical palliation of a symptomatic primary.A significant proportion also DOI:http://dx.doi.org/10.7314/APJCP.2014.15.15.6059 Descriptive Epidemiology of Colorectal Cancer in UMMC, 2001to 2010 received chemotherapy in the form of adjuvant therapy, and a few had radiotherapy depending on the tumor location, stage and general wellbeing of the patient; this reflects the late stage of presentation of our patients.
Our study has several strengths, including the use of a large sample of patients, from a tertiary hospital, with a functional multi-disciplinary team dedicated to the care of colorectal cancers.The exclusion of non-Malaysian patients is useful as we want to identify any environmental associations.The other point is that our data encompasses a longer time frame than any of the registry data, and has more clinical treatment data.
Apart from the strengths listed above, some limitations need to be highlighted.One limitation is in the retrospective component of the study.Patients were identified using ICD codes from the hospital medical records.This coupled with the lack of synoptic histopathology reporting and other patient details in the early years (before 2008) limit the completeness of data.Our finding therefore, needs verification in larger population-based dataset.
In conclusion, systematic data collection monitoring of patients follow up to ensure quality data are obtained for both hospital based studies should be undertaken.In addition, synoptic reporting of histopathology examinations is urgently needed.At the same time, for any measure control and preventive measures to work, the characteristics and differences identified among colorectal cancer patients must be considered.For example, health education and screening campaigns should target both men and women equally, and screening should be carried out at younger age for the Malays compared to the Chinese and Indians.
This study adds to the scant body of knowledge regarding characteristics of patients with colorectal cancer, and thus provides a foundation for further studies on the subject matter.Further research is needed to clarify the risk factors of colorectal cancer among the different ethnic groups in Malaysia, the role of tumor biology, genetics, and lifestyle risk factors in explaining differences in how colorectal cancers present and behave.

Figure 1 .
Figure 1.Tumor Grade Distribution by Age Groups

Table 4 . Treatment Modalities, Operation Types and Patients Presentation
*Note. a Anterior resection including high, low and ultralow anterior resection; b Abdomino Perineal Resection; c Right hemi colectomy including (extended right); d Left hemi colectomy including (extended left) and; e Other procedures e.g.appendectomy 862 patients.As shown in Figure