Pancreatic Cancer And Chronic Pancreatitis Investigated By Nuclear Magnetic Resonance Spectroscopy

Pancreatic cancer(also called pancreatic carcinoma)is a malignant tumor of the pancreas,which accounts for only 2%of all cases of cancer worldwide,but it has the worst prognosis of all cancers,with a 5-year survival rate of＜3%.Because of the deep location and unnoticeable symptoms,pancreatic cancer often starts silently without symptoms at its early stage but progresses quite rapidly and has a poor prognosis.The medical community holds that the best cure is resection at the early stage.The clinical diagnosis of pancreatic cancer is highly dependent on imaging examination,in which magnetic resonance imaging(MRI)is a valuable mean for measuring and imaging metabolic processes in vivo,and is also indispensable in confirming and locating pancreatic cancer.However,the most troublesome problem in MRI diagnosis was always to differentiate pancreatic cancer from chronic pancreatitis.This paper was applied nuclear magnetic resonance(NMR)-metabonomics approach,combined NMR technique with pattern recognition analysis(PCA)method, to evaluate the feasibility of applying NMR-based metabonomic approach to earlier detect the differences among pancreatic cancer,chronic pancreatitis,and the healthy controls.It was designed to launch at two different aspects,animal models and clinical cases.The experimental procedures are as follows:1.the metabolic profiles of pancreatic juice and tissue samples were investigated by NMR;2.NMR data obtained were processed by PCA for multivariate statistics analysis;3.the results from animal models and clinical cases,respectively,were integrated to set up a system NMR research data.In the animal model,the different metabolic profiles between chronic pancreatitis and pancreatic cancer in rats relative to normal pancreas were studied in vitro by high-resolution magic angle spinning nuclear magnetic resonance(HRMAS NMR) spectroscopy,using 1D ~1H single-pulse and CPMG and 2D J-resolved,COSY, TOCSY and HSQC pulse sequences.PCA was then utilized to facilitate automated NMR pattern recognition in the form of scores and loading plots.In the clinical eases,liquid-state NMR was applied to investigate pancreatic juice of patient suffered with chronic pancreatitis and pancreatic cancer.The 1D NOESY and 2D J-resolved,COSY,and TOCSY NMR experiments were carried out and the main resonances were assigned.Meanwhile,the specimens of pancreatic cancer tissue, including low-differentiation,middle-differentiation,and high-differentiation cancer, and adjacent normal pancreas parts excised from the patients were investigated by HRMAS NMR combined with PCA.Both the NMR experiments and PCA methods were similar with those in the animal model except 1D NOESY was chosen in place of single-pulse.The relationship between the metabolic changes and the findings of histopathology was also studied.It's encouraging that the results from the rat model and the patient cases were highly consistent:compared to the normal group,pancreatic cancer group represented elevated leucine,iso-leucine,valine,lactate,and taurine,accompanied with depleted betaine,phosphocholine and glycerophosphocholine levels.Besides,more valuable findings were obtained:the signals of 3-hydroxy- butyric acid and cholesterol appeared only in the pancreatic cancer tissues,and the fluctuated changes of inositol were obvious only in human pancreatic cancer.While relatively increasing phosphocholine and glycerophosphocholine and decreasing leucine,iso-leucine, valine,and lactate were observed in chronic pancreatitis in animal model.The results of pancreatic juice showed that ~1H NMR spectra of chronic pancreatitis and pancreatic cancer groups were very similar except of the resonances of ethoxyl,which were observed only in chronic pancreatitis group.In the scores plots resulted from the progressive PCA process to the ~1H NMR data,the great variability between the normal,pancreatic cancer and/or pancreatitis group and was presented.Based on the above integrated results,the paper suggests that elevated lactate, 3-hydroxybutyric acid,taurine and betaine and especially depleted phosphocholine and glycerophosphocholine be the metabolite biomarkers for pancreatic cancer.On the exploitation of the NMR-metabonomics research on intact pancreatic cancer tissues from both animals and patients,the potential of HRMAS NMR spectroscopy combined PCA to discriminate metabonomic profile of pancreatic cancer from normal pancreas was verified.The present data might contribute to a better knowledge of the abnormal metabolism of pancreatic carcinogenesis and a valuable assistance for MRI diagnosis and assessment of therapeutic response.However,the greater number of samples and further research is needed in this project to reach a high degree of precision and reproductivity.