Thesis (Ph. D.)--Washington State University, May 2006. / Includes bibliographical references.

Thesis (M.Phil.)--Hong Kong University of Science and Technology, 2007. / Includes bibliographical references (leaves 136-157). Also available in electronic version.

No description available.

No description available.

Background: Pancreatic disease is a global problem. Severe acute pancreatitis (AP) carries a 30-50% mortality. Current scoring systems fall short in predictive accuracy, sensitivity, specificity and availability. Pancreatic cancer (PC) is a leading cause of cancer-related mortality, most patients die within one year of diagnosis. Late presentation and lack of effective oncological treatment determine a desperate need to focus on early detection of the pancreatic cancer. Current biomarkers fall short in accessibility, sensitivity and specificity and ability to distinguish malignant from benign conditions. Metabolomics aims to decipher molecular signatures that will distinguish disease from controls, ultimately leading to novel targets for diagnosis and treatment. Initial studies are discovery-based, hypothesis-generating and typically aim to establish a snapshot of the metabolism of an individual by metabolite profile. Aims: Establish a prospective phenotypic and demographic database of patients with acute pancreatitis. Determine urinary and serum metabolomic profiles of AP and PC in comparison to controls and establish if metabolomic profiling can distinguish severity of each disease in order to identify potential novel bio-markers. Methods: Urine and serum samples from 73 AP, 32 PC, 62 Healthy Controls, 8 Chronic pancreatitis and 8 Benign jaundice participants were analysed using GC-MS and UPLCMS. Metabolite identification was subject to univariate and multivariate analysis (p<0.05). Results: The differentiation of metabolite profiles was most distinct with AP. There was no differentiation by AP aetiology. AP severity was distinquished by metabolite profile. Profiles of resectable patients were distinct form non-resectable PC. Fatty acids(FA), glycerophoshocholines, eicosanoids, TCA cycle intermediates and melatonin levels were altered in AP. PC was defined by altered concentrations of FAs, eicosanoids, glycerophoshocholines, sphingomyelins, folates and amino acids and peptides (e.g. glutamine). Altered levels of UFAs, neuromedins, Vitamin D3 determined stage of PC. Conclusion: Urinary and serum metabolomic signatures may provide future biomarker panels for grading AP and PC.

Background: The use of novel targeted anticancer drugs in clinical practice is rapidly increasing. As the use of these drugs increases, so does the need to develop biomarkers to optimise the drugs' clinical and cost effectiveness. The attitudes and views of all stakeholders regarding the optimal use of predictive biomarkers in guiding personalised medicine are crucial for identifying acceptable criteria of predictive biomarker tests to guide future biomarker development. To gain insight into these views, we aimed to develop and validate a survey tool that would aid in assessing attitudes of cancer physicians and patients regarding the utilisation of biomarkers in tailoring treatment according to individual patient needs. Methods: Two questionnaires (one for oncologists and one for patients) based on emerging clinical data about novel targeted anticancer agents were designed to capture information about acceptable sensitivity, specificity, invasiveness and cost of a predictive biomarker test. A hypothetical scenario was provided that described a newly developed, targeted anticancer drug that was found to be more beneficial to certain patient subgroups identified through a predictive biomarker test. Questions in the patient survey were based on the results of the oncologist survey. Results: The response rates to these surveys were 20% (n=67) for the oncologists and 59% (n=100) for the patients. The oncologists' attitudes regarding the predictive biomarker test's false negative (FN) and false positive (FP) rates varied with the level of health outcome generated by the hypothetical drug. The acceptable FN rate for predictive biomarker test results detected in the current study was similar to many current predictive biomarkers, but the FP rate considered acceptable was much lower. The majority of the patients (90%) accepted the median acceptable FN rate of 10% reported by the oncologists. A significant minority of the oncologists (27%) refused a tumour biopsy (in addition to the diagnostic biopsy) to collect samples for the purpose of predictive biomarker testing. A much higher percentage of patients (68%) refused a biopsy under such conditions. Interestingly, our data also suggest that oncologists' expectations for the outcome of therapy have changed little in recent years, while patients' expectations have increased dramatically. Conclusions: Our data suggest that oncologists and patients agree that a FN rate of 10% is acceptable. However, based on the oncologists' responses, the FP rates associated with current predictive biomarkers are far from ideal. This may reflect oncologists' pragmatic approach in the absence of alternative choices for predictive biomarker tests. However, it also suggests that future biomarker development and implementation must focus on decreasing the FP rate without increasing the FN rate. Recent results demonstrating molecular heterogeneity in tumours suggest that, considering our data on acceptable test accuracy, serial/repeated tumour biopsies may be required. However, based on the attitudes of physicians and patients reported here, such ‘biopsy-driven' clinical trials may not be acceptable, and, without further investigation or education, this may be a barrier to the successful implementation of such potentially valuable investigational strategies. These results should certainly be taken into account when planning biopsy-dependent trials, and emphasises the importance of pursuing non-invasive biomarker assays. We believe that the survey questionnaires originating from the current study are valid tools for assessing stakeholders' attitudes and views about the appropriate application of predictive biomarkers in personalised medicine.

No description available.

published_or_final_version / abstract / Zoology / Doctoral / Doctor of Philosophy

published_or_final_version / Surgery / Master / Master of Philosophy

published_or_final_version / Surgery / Doctoral / Doctor of Philosophy