Colorectal cancer survival in Jordan 2003-2007

Al Nsour, Mohannad (2014) Colorectal cancer survival in Jordan 2003-2007. PhD thesis, University of Glasgow.

Abstract

As in other less developed countries in the region and elsewhere, cancers are becoming a major cause of morbidity and mortality in Jordan. Globally, colorectal cancer (CRC) is the second most common cancer in women and the third most common cancer in men. In Jordan, CRC is the second most common cancer in women and the most common in men. There is little known about survival from CRC in Jordan and few survival studies have been conducted in comparable Eastern Mediterranean countries.

As the first study of its type in Jordan, this thesis aimed at estimating CRC survival among Jordanian patients and comparing them with survival estimates among other populations. The thesis explored the relationship between CRC and socio-demographic characteristics, clinical manifestations, treatment, diabetes mellitus – for which the prevalence in Jordan is very high - and treatment sites. The study augmented existing Jordan Cancer Registry data by gathering additional case mix information and completing missing fields. CRC was classified according to international classification of oncology (ICD-O third edition in addition to the International Classification of Diseases, ICD-10) as C18.0-C20.9. The vital status of the patients was ascertained from Civil Registration Bureau based on use of the unique National Identification number of the patients with follow-up to 31 December, 2010. The survival duration of each case was determined as the time difference (in days) between the date of incidence (index date) and the date of death, date of loss to follow-up or the closing date for follow-up (31 December, 2010). Observed and relative CRC survival rates were calculated among a study population of 1,896 Jordanian colorectal patients aged 15 to 99 years of age, diagnosed with first invasive primary CRC from January 2003 to December 2007. The Kaplan-Meier method was used to determine the observed survival probability over time. The logrank test was used to estimate whether the difference in survival estimates was statistically significant between the groups. The complete approach of estimation of observed survival probability was used. Cox proportional hazard regression was used to assess the effect of each variable after simultaneously controlling the effects of potential confounders.

With half of the sample aged 60 years and above, males were predominant (55.5 percent) with the majority of the sample (75.4 percent) residing in the central part of Jordan. The vast majority of the cases (63.5 percent) were diagnosed with colon cancer, with regional metastasis present in 58.9 percent. No significant difference was found in the distribution of colon and rectum cancer by sex. Adenocarcinoma was the most commonly found tumor (84.4 percent) compared to mucinous tumors which was found in 7.8 percent of the patients. In addition, 62.7 percent of the cases were classified as moderate and 14.9 percent as poor. The percentages of rectum cancer patients with moderate and poor/anaplastic types of cancer were higher than in colon cancer patients. The majority of patients (77.9 percent) underwent surgery, which was mostly elective (82.0 percent). A significantly higher occurrence of these elective surgeries was found among rectum cancer (87.7 percent) than colon cancer patients (78.7 percent). Curative treatment was found to be a more common form of treatment for colorectal cancer patients (76.5 percent) than palliative (23.5 percent). Of those undergoing surgery, 4.8 percent has died within 30-days of resection, with a significantly lower mortality among patients aged ≤ 65 years (2.9 percent) than the over 65 years age group (7.1 percent). Thirty days postoperative mortality was significantly higher among colon cancer patients (5.3 percent), patients with more advanced tumours and those who underwent emergency operations.
Results of this study revealed that the incidence of CRC in the Jordanian population to be low compared to developed countries. However, this low incidence is similar to CRC incidence rates in other countries in the region. During the 5-year study period, the overall crude colorectal cancer incidence rate for males was 5.6 per 100,000 population, and 5.1 per 100,000 populations in females. The overall Age Standardized colorectal incidence rate (ASR) among males was 15.5 per 100,000 populations compared to 12.5 per 100,000 populations among females. For colon cancer, the crude incidence rate was 5.4 per 100,000 populations in males and 4.1 per 100 000 populations in females, while ASR for males was 11.1 and 8.4 for females. Alternatively, the crude incidence rate for rectum cancer was 3.0 per 100,000 population for males and 2.4 per 100,000 population for females, and the respective ASR incidence rates was 6.1 per 100,000 males and 4.9 per 100,000 females. Unexpectedly, results showed a high percentage (13.8 percent) of CRC patients among the young age groups (i.e. less than 40 years of age) with insignificant differences between the sexes. The age specific incidence rates were found to increase with age.

The study revealed that 5-year observed and relative survival probabilities for colorectal cancer to be 57.7 and 61.3 percent respectively, with higher probability for colon cancer. These results showed good survival estimates of colorectal cancer compared to developed countries as well as the most developed countries in the region and across the Asian continent. The slightly higher observed colorectal survival rates among females were found to be insignificantly different than those for males. Patients aged 45 through 59 years had the highest survival estimates among all age groups, and the 75 years and above age group the lowest. The highest survival estimates were found among patients living in the central parts of Jordan, and the poorest was significantly noted in the south. Moreover, the observed and relative survival estimates were consistently highest during Year 1 and lowest during Year 5.

Mucinous and serous tumors showed the poorest survival rates among the colorectal cancer, with higher 5-year relative survival rates among the mucinous and serous type of colon (52.4 percent) compared to rectum cancer (42.8 percent). With more than half of the colon patients (57.2 percent) and rectal patients (62 percent) having a regional spread; a higher proportion of colon cancers (24.1 percent) were found to have distant metastasis, than rectum cancer patients (20.5) and an equally low (11 percent) had localized CRC at diagnosis. Results also showed that observed and relative survival rates from localized and regional colon cancer were better than survival from rectum cancer in the same stages at 1, 3 and 5-years of diagnosis. Results also indicated that observed survival became poorer with increasing age for both localized and regional tumours. This observation was applicable for both males and females.

In recognizing appropriate surgery as the most important aspect of colorectal cancer treatment, the observed survival probability for colorectal cancer patients who underwent surgery was found significantly higher than that for patients who did not undergo surgery. This scenario was similarly observed for both colon and rectum. Conversely, the overall relative survival rates for patients who underwent surgery declined from 96.2 percent to 62.6 percent between the first and fifth year and from 86.5 percent to 23.5 percent for patients who did not undergo surgery. In addition, no significant difference was found between colorectal survival estimates for patients who underwent elective surgery and those who underwent emergency surgery. This was held true for both colon and rectum cancer.

A multivariable analyses done to examine the adjusted odds of death within 30 days of surgery and selected variables revealed that the odds of dying were significantly higher among colorectal cancer patients older than 65 years (OR 2.3, 95percent CI: 1.3-4.1), those with distant tumors (OR 3.6, 95percent CI: 2.0-6.2); and those who were operated upon as an emergency (OR 2.3, 95percent CI: 1.2-4.1).

Study findings indicated that colorectal cancer patients who received chemotherapy treatment had better survival for almost the first four years. However this was not a statistically significant result. Similarly, colon cancer patients who received chemotherapy treatment had better survival rates for nearly four years, compared to rectum cancer patients had better survival rates for the first two years from receiving chemotherapy.

In terms of treatment sites (hospitals), results showed that 32.4 percent of cases were treated at public health facilities, 23.4 percent at King Hussein Cancer Center (KHCC), 18.4 percent at the teaching hospitals, 16.8 percent at the private health facilities, and only 9.1 percent at other sites. The results of Cox proportional hazards ratios, after adjusting for age, extent of disease, place of residence, surgery, chemotherapy, radiotherapy and intent of treatment, have shown that patients who received treatment in private hospitals as well as in the King Hussein Cancer Center, had better survival rates compared to those who received their treatment in the public sector.
Finally, this study revealed that the mean survival for colorectal cancer patients with diabetes mellitus was significantly lower than that for patients without diabetes (Log-Rank test, p=0.0359). The study also revealed a significant relationship between diabetes mellitus and colon cancer survival, where diabetic patients with colon cancer were less likely to survive compared to non-diabetic patients with colon cancer. However, no significant association was observed regarding diabetic patients with rectum cancer. In addition, multivariate analysis identified diabetes mellitus as a significant predictor associated with lower observed survival, where diabetic patients were one and one-half times more likely to be at risk of dying compared to non-diabetic patients. Age group 75 years or older, regional and distant metastasis of disease were shown to be independent prognostic factors for observed survival in this multivariate analysis.

The Cox proportional hazard model showed that age, place of residency, extent of disease, and morphology to be significant predictors for colorectal cancer survival. Colon cancer patients aged 75 years and above had 2.2 times higher risk of death than those aged 44 years or less ( HR=2.2, 95percent CI: 1.5-3.1). Rectum cancer patients residing in the central region of the country had a 27 percent lower risk of death compared with those who resided in the North (HR=0.73, 95percent CI: 0.56-0.95). Colon and rectum cancer patients with regional metastasis had three times and one and the half times higher risk of death than those with localized disease respectively (Colon: HR=3.3, 95percent CI: 2.0-5.6); (Rectum: HR=1.6, 95percent CI: 1.1-2.5). Moreover, colon patients with distant metastasis had fourteen times higher risk of death and rectum cancer patients portrayed four and the half times higher risk of death than those with localized disease (Colon: HR=14.0, 95percent CI: 8.0-23.8); (Rectum: HR=1.6, 95percent CI: 2.8-7.5). Colon patients with poor or anaplastic grade had almost twice the risk of death than those with well grade (HR=1.9, 95percent CI: 1.1-3.2). On the other hand, patients diagnosed with mucinous rectum cancer had 1.4 times higher risk of death than those with adenocarcinoma (HR=1.4, 95percent CI: 1.1-2.1). Sex, grade and year of diagnosis were insignificant predictors across all three models (CRC, colon and rectum).

The main highlights of the study findings included a higher than expected incidence and a poorer than expected colorectal survival among the younger population, a large variation in survival rates based on the place of residency, and a significantly poorer survival among diabetic patients. Results of this study prompted a set of recommendations to assist national efforts in preventing and improving the survival of colorectal cancer in Jordan. These recommendations included various actions and measures to strengthen health service provision; assure provision of health care by expanding services; improve monitoring by promoting policy and research; and improve and strengthen data quality measures. Specifically, the study recommended screening to be made at a younger age in Jordan than in other developed countries as a result of the high percentage of CRC (13.8percent) among young age groups (<40 year) that was reported in the study. Conducting further research to investigate the reasons for poor survival rates among younger CRC patients is presented as a strong study recommendation.

In addition, introducing managed clinical networks as an approach for reducing the variation in survival between the different hospitals was presented as a worthwhile recommendation. Undoubtedly, improving public health efforts to reduce the prevalence of Diabetes Mellitus, in addition to undertaking further research to explain the increased mortality among diabetic colorectal patients are presented as valuable recommendations in this study. Finally, the study recommended that the Jordan Cancer Registry play a major role in following-up with cancer patients to examine the quality of cancer services that they receive.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: Cancer, colorectal, incidence, survival, Jordan
• Subjects: R Medicine > R Medicine (General); R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Health & Wellbeing > Public Health
• Supervisor's Name: Morrison, Dr David and Forman, Dr David
• Date of Award: 2014
• Depositing User: Dr Mohannad Al Nsour
• Unique ID: glathesis:2014-5063
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 06 Aug 2014 10:22
• Last Modified: 31 Aug 2015 08:49
• URI: https://theses.gla.ac.uk/id/eprint/5063