Doctor of Philosophy / Introduction Cachexia is characterised by a marked weight loss and the presence of anorexia, anaemia, and asthenia. Although cachexia is often associated with the presence and growth of tumour and observed in solid tumours of the upper gastrointestinal tract, its presence is not unique to cancer and is often also present in most chronic, end-stage diseases processes. The loss of body fat, altered lipid metabolism, increase in the resting energy expenditure, and the increased loss of body protein the degree of which is associated with poor survival, are all hallmarks of this detrimental disease. The clinical aspects and consequences of cachexia can simply be summarised as morbidity, debilitating conditions, and mortality. The conditions such as loss of muscle mass, impaired muscle function, fatigue, reduced activity and functional capacity by themselves are enough to severely and significantly affect the patients’ QL. Although different interventional procedures and therapies are available for the treatment of cachexia and its symptoms, effective methods to evaluate their benefits and outcomes have not been tested or investigated. It was, therefore, the aim of this project to use body composition analysis as a clinical tool and evaluate the effectiveness and outcome of interventional and therapeutic procedures in three groups of patients with cancer. Methods Three patient groups were investigated: 1) patients with pancreatic cancer undergoing Whipple’s Procedure, 2) patients with pancreatic cancer undergoing cancer chemotherapy and receiving either EPA or placebo, and 3) patients with malignant mesothelioma undergoing cancer chemotherapy plus thalidomide or thalidomide alone. Body composition analysis techniques were used to assess the changes in TBN, TBF, TBK, and TBW. In addition, the body composition parameters together with clinical measures were also used to determine parameters influencing survival. The malignant mesothelioma patients were randomised into patients who received gemcitabine / cisplatin plus thalidomide and those who received thalidomide alone. The pancreatic cancer patients undergoing chemotherapy were randomised into the group who were receiving EPA and those who were receiving placebo. In addition, these patients were also investigated on the basis of their disease extent where they were separated into two groups of metastatic and locally advanced. Unpaired T-Test and ANOVA were used to determine differences between groups. Kaplan-Meier analysis and Cox’s Regression were used to assess survival in all three patient cohorts. The Whipple’s Procedure patients were separated into those who received a Clear Margin and those who received an Unclear Margin during their resection. Results 1) In the pancreatic cancer patients undergoing Whipple’s Procedure, compared to the base-line, there were highly significant changes in Weight (p=0.006), BMI (p=0.005), and FM (p=0.007) followed by significant changes in %BFat (p=0.016), TBK/Ht (p=0.021), LBM (By TBK) (p=0.023), LBM (Van Loan) (p=0.034), and LBM (Segal) (p=0.038) at the 14 week time-point. At the 26 weeks post-operative time point, the only significant changes were in the FM (p=0.012), %BFat (p=0.003), and BMI (p=0.027) parameters. There was also a deviation between the two groups in their TBN, LBM and TBW content observable in a long-term setting and fat content in the relatively shorter-term. Although the Unclear Margin group had lower body composition values, both groups seem to begin to gradually “equalise” around the 14 weeks post-operative time-point. The survival analysis results for the Whipple’s Procedure patients demonstrated that Margin Status (p=0.001), Fat Mass (p=0.003) and Age (p=0.081) were significant and could influence survival. 2) When the second cohort pancreatic cancer patients undergoing chemotherapy were analysed, they were initially separated according to the extent of their disease The results of the analyses of body composition changes between measurement time-points for the each group separately, suggested that the patients with locally advanced disease maintain their Weight, FM, and TBN but are more likely to have a lower TBW by the end of the four month of chemotherapy. However, the patients with metastatic pancreatic cancer maintain their TBW but are more likely to have a decreased fat compartment and a higher FFM. The QL analysis showed that the metastatic group are performing “worse” than the locally advanced group especially in term of their Dyspnoea, Nausea & Vomiting, and Sexuality. In addition, the Karnofsky score showed that the metastatic group are not performing as well as the locally advanced group. Furthermore, for the metastatic group there was an increase in the patients’ pain with a decline in mood and general performance as well as increase in gastrointestinal symptoms. Pain Card scores also showed a general increase for the metastatic group and a general decrease for the locally advanced group. When the pancreatic cancer patients undergoing chemotherapy were separated according to whether they received EPA or placebo, the results demonstrated that firstly, due to the fact that the patients were well randomised, the two groups commenced the trial with similar and statistically non-significantly different body composition parameters. Secondly, the two groups were also found to be statistically not different at their corresponding measurement time-points. And thirdly, the patients receiving placebo compared to those receiving EPA lost more Weight, and FM but less TBW throughout the trial. The TBK/Ht (p=0.044), TBK (p=0.042), and LBM (By TBK) (p=0.042), however, showed statistically significant differences where in all three parameters the EPA showed an increase compared to the base-line (pre-chemotherapy). Results of the survival analysis demonstrated that the use of EPA in this group of pancreatic cancer patients did not provide any benefit. In fact, as it was shown in the Kaplan-Meier plot, the group of patients receiving the EPA had a “worse” survival than the group receiving the placebo. The QL results showed that placebo group improved in their functional scales, but increased their Altered Bowel Habit scores with an increase in the perception of pain and decrease in relief from pain. The EPA group, however, showed a decrease in the Loss of Appetite, Dyspnoea, Pain, Pancreatic Pain, and Fatigue, and improvements in Role Functioning and Sexuality. 3) Results of the malignant mesothelioma patients demonstrated that both study arms show similar weight changes. In addition, body composition measurements indicated that the gemcitabine / cisplatin chemotherapy plus thalidomide group had a greater TBN loss and a greater TBW gain than the thalidomide-alone group. This loss of TBN and gain in TBW looked to be “concealed” in the weight. The results of the survival analysis carried out on the mesothelioma patient group suggested that haemoglobin levels (p=0.001), Age (p=0.007), and NI (p=0.008) are the parameters that can influence the survival of patients with malignant mesothelioma undergoing chemotherapy. Conclusions 1) The trend in body composition changes in the Whipple’s Procedure group showed that, although both groups may start with non-significantly different body composition, they tended to grow closer around the 14 week point indicating that the Clear Margin group may lose more than Unclear Margin group. The implications of these findings, therefore, were that once the most appropriate surgical procedure is performed, an adjuvant therapy regimen (such as chemotherapy) at around 14 weeks may have the most impact on the patient’s overall treatment outcome. 2) When the pancreatic cancer patients were separated by the extent of their disease, the results lead to the conclusion that the patients with locally advanced disease maintain their Weight, FM, and TBN but are more likely to have a lower TBW by the end of the four month of chemotherapy. However, the patients with metastatic pancreatic cancer maintain their TBW but are more likely to have a decreased fat compartment and a higher FFM. The QL analysis concluded that the results may point to a worsening and/or progressing disease which is consistent with classic metastatic disease aetiology. From the results of the pancreatic cancer patients undergoing cancer chemotherapy it was concluded that the use of EPA in this group of pancreatic cancer patients undergoing cancer chemotherapy with gemcitabine results in a non-significant reduction in weight loss, FM loss, and TBW gain with a statistically significant increase in FFM. The results of the survival analysis was, however, contradictory suggesting that patients receiving EPA may have a worse survival than the placebo group. The QL analysis here concluded that that EPA does improve the QL of this group of pancreatic cancer patients. 3) From the malignant mesothelioma group it was concluded that provided that the overall anti-cancer potential of gemcitabine / cisplatin plus thalidomide is comparable with that of thalidomide-alone, then by looking purely from the body composition angle one may be able to suggest the use of thalidomide alone in the treatment of malignant mesothelioma in this group of patients. From the results of the survival analysis, the fact that the Study Arm parameter did not reach statistical significance could indicate that survival in these patients is not affected by the presence or absence of chemotherapy with gemcitabine and cisplatin. The body composition techniques were used here as a tool to monitor changes in various body composition parameters to assess the outcomes, including survival, of the administration of different therapies and interventional procedures in these three groups of cancer patients. For these purposes, these techniques were demonstrated to be an effective and invaluable tool.

Doctor of Philosophy / Introduction Cachexia is characterised by a marked weight loss and the presence of anorexia, anaemia, and asthenia. Although cachexia is often associated with the presence and growth of tumour and observed in solid tumours of the upper gastrointestinal tract, its presence is not unique to cancer and is often also present in most chronic, end-stage diseases processes. The loss of body fat, altered lipid metabolism, increase in the resting energy expenditure, and the increased loss of body protein the degree of which is associated with poor survival, are all hallmarks of this detrimental disease. The clinical aspects and consequences of cachexia can simply be summarised as morbidity, debilitating conditions, and mortality. The conditions such as loss of muscle mass, impaired muscle function, fatigue, reduced activity and functional capacity by themselves are enough to severely and significantly affect the patients’ QL. Although different interventional procedures and therapies are available for the treatment of cachexia and its symptoms, effective methods to evaluate their benefits and outcomes have not been tested or investigated. It was, therefore, the aim of this project to use body composition analysis as a clinical tool and evaluate the effectiveness and outcome of interventional and therapeutic procedures in three groups of patients with cancer. Methods Three patient groups were investigated: 1) patients with pancreatic cancer undergoing Whipple’s Procedure, 2) patients with pancreatic cancer undergoing cancer chemotherapy and receiving either EPA or placebo, and 3) patients with malignant mesothelioma undergoing cancer chemotherapy plus thalidomide or thalidomide alone. Body composition analysis techniques were used to assess the changes in TBN, TBF, TBK, and TBW. In addition, the body composition parameters together with clinical measures were also used to determine parameters influencing survival. The malignant mesothelioma patients were randomised into patients who received gemcitabine / cisplatin plus thalidomide and those who received thalidomide alone. The pancreatic cancer patients undergoing chemotherapy were randomised into the group who were receiving EPA and those who were receiving placebo. In addition, these patients were also investigated on the basis of their disease extent where they were separated into two groups of metastatic and locally advanced. Unpaired T-Test and ANOVA were used to determine differences between groups. Kaplan-Meier analysis and Cox’s Regression were used to assess survival in all three patient cohorts. The Whipple’s Procedure patients were separated into those who received a Clear Margin and those who received an Unclear Margin during their resection. Results 1) In the pancreatic cancer patients undergoing Whipple’s Procedure, compared to the base-line, there were highly significant changes in Weight (p=0.006), BMI (p=0.005), and FM (p=0.007) followed by significant changes in %BFat (p=0.016), TBK/Ht (p=0.021), LBM (By TBK) (p=0.023), LBM (Van Loan) (p=0.034), and LBM (Segal) (p=0.038) at the 14 week time-point. At the 26 weeks post-operative time point, the only significant changes were in the FM (p=0.012), %BFat (p=0.003), and BMI (p=0.027) parameters. There was also a deviation between the two groups in their TBN, LBM and TBW content observable in a long-term setting and fat content in the relatively shorter-term. Although the Unclear Margin group had lower body composition values, both groups seem to begin to gradually “equalise” around the 14 weeks post-operative time-point. The survival analysis results for the Whipple’s Procedure patients demonstrated that Margin Status (p=0.001), Fat Mass (p=0.003) and Age (p=0.081) were significant and could influence survival. 2) When the second cohort pancreatic cancer patients undergoing chemotherapy were analysed, they were initially separated according to the extent of their disease The results of the analyses of body composition changes between measurement time-points for the each group separately, suggested that the patients with locally advanced disease maintain their Weight, FM, and TBN but are more likely to have a lower TBW by the end of the four month of chemotherapy. However, the patients with metastatic pancreatic cancer maintain their TBW but are more likely to have a decreased fat compartment and a higher FFM. The QL analysis showed that the metastatic group are performing “worse” than the locally advanced group especially in term of their Dyspnoea, Nausea & Vomiting, and Sexuality. In addition, the Karnofsky score showed that the metastatic group are not performing as well as the locally advanced group. Furthermore, for the metastatic group there was an increase in the patients’ pain with a decline in mood and general performance as well as increase in gastrointestinal symptoms. Pain Card scores also showed a general increase for the metastatic group and a general decrease for the locally advanced group. When the pancreatic cancer patients undergoing chemotherapy were separated according to whether they received EPA or placebo, the results demonstrated that firstly, due to the fact that the patients were well randomised, the two groups commenced the trial with similar and statistically non-significantly different body composition parameters. Secondly, the two groups were also found to be statistically not different at their corresponding measurement time-points. And thirdly, the patients receiving placebo compared to those receiving EPA lost more Weight, and FM but less TBW throughout the trial. The TBK/Ht (p=0.044), TBK (p=0.042), and LBM (By TBK) (p=0.042), however, showed statistically significant differences where in all three parameters the EPA showed an increase compared to the base-line (pre-chemotherapy). Results of the survival analysis demonstrated that the use of EPA in this group of pancreatic cancer patients did not provide any benefit. In fact, as it was shown in the Kaplan-Meier plot, the group of patients receiving the EPA had a “worse” survival than the group receiving the placebo. The QL results showed that placebo group improved in their functional scales, but increased their Altered Bowel Habit scores with an increase in the perception of pain and decrease in relief from pain. The EPA group, however, showed a decrease in the Loss of Appetite, Dyspnoea, Pain, Pancreatic Pain, and Fatigue, and improvements in Role Functioning and Sexuality. 3) Results of the malignant mesothelioma patients demonstrated that both study arms show similar weight changes. In addition, body composition measurements indicated that the gemcitabine / cisplatin chemotherapy plus thalidomide group had a greater TBN loss and a greater TBW gain than the thalidomide-alone group. This loss of TBN and gain in TBW looked to be “concealed” in the weight. The results of the survival analysis carried out on the mesothelioma patient group suggested that haemoglobin levels (p=0.001), Age (p=0.007), and NI (p=0.008) are the parameters that can influence the survival of patients with malignant mesothelioma undergoing chemotherapy. Conclusions 1) The trend in body composition changes in the Whipple’s Procedure group showed that, although both groups may start with non-significantly different body composition, they tended to grow closer around the 14 week point indicating that the Clear Margin group may lose more than Unclear Margin group. The implications of these findings, therefore, were that once the most appropriate surgical procedure is performed, an adjuvant therapy regimen (such as chemotherapy) at around 14 weeks may have the most impact on the patient’s overall treatment outcome. 2) When the pancreatic cancer patients were separated by the extent of their disease, the results lead to the conclusion that the patients with locally advanced disease maintain their Weight, FM, and TBN but are more likely to have a lower TBW by the end of the four month of chemotherapy. However, the patients with metastatic pancreatic cancer maintain their TBW but are more likely to have a decreased fat compartment and a higher FFM. The QL analysis concluded that the results may point to a worsening and/or progressing disease which is consistent with classic metastatic disease aetiology. From the results of the pancreatic cancer patients undergoing cancer chemotherapy it was concluded that the use of EPA in this group of pancreatic cancer patients undergoing cancer chemotherapy with gemcitabine results in a non-significant reduction in weight loss, FM loss, and TBW gain with a statistically significant increase in FFM. The results of the survival analysis was, however, contradictory suggesting that patients receiving EPA may have a worse survival than the placebo group. The QL analysis here concluded that that EPA does improve the QL of this group of pancreatic cancer patients. 3) From the malignant mesothelioma group it was concluded that provided that the overall anti-cancer potential of gemcitabine / cisplatin plus thalidomide is comparable with that of thalidomide-alone, then by looking purely from the body composition angle one may be able to suggest the use of thalidomide alone in the treatment of malignant mesothelioma in this group of patients. From the results of the survival analysis, the fact that the Study Arm parameter did not reach statistical significance could indicate that survival in these patients is not affected by the presence or absence of chemotherapy with gemcitabine and cisplatin. The body composition techniques were used here as a tool to monitor changes in various body composition parameters to assess the outcomes, including survival, of the administration of different therapies and interventional procedures in these three groups of cancer patients. For these purposes, these techniques were demonstrated to be an effective and invaluable tool.