Functional Molecular Imaging In Hepatology

The liver plays a vital role in the metabolism of endogenous and exogenous substances. Its location between the gastrointestinal tract and the systemic circulation means that food and other consumed substances have to pass through the liver before reaching the systemic circulation. The liver receives blood from the gut via the portal vein (approx. 75%) and from the hepatic artery (approx. 25%). The blood from these vessels mixes at the entrance to the liver sinusoids which are specialised liver capillaries lined by highly fenestrated endothelial cells through which the majority of substrates can pass directly from the blood to the hepatocytes via an extended plasma volume, the space of Dissé. The vascular structure of the liver is thus optimal for the exchange of substances with the blood. From the sinusoids, the blood flows into the liver veins and is subsequently returned to the systemic circulation. The vascular structure and the interplay between its components is dynamic and of essential importance for the function of the liver.

This chapter provides an overview of methods for the measurement of hepatic blood flow for the whole liver and hepatic blood perfusion per unit volume of liver tissue. Measurement of the total flow rate is performed by intravenous infusion of a test substance that is eliminated only by the liver, measurements of steady-state arterial and liver venous concentration differences across the organ and calculation by Fick’s principle. Indocyanine green (ICG) has been validated as a suitable test substance for such measurements in studies using pigs and is widely used in human studies. Steady-state clearance measurements can be used to measure the total hepatic flow rate for substances with a liver vein concentration that is negligible when compared to the arterial concentration. Sorbitol has been validated as a suitable test substance in healthy subjects (with corrections for urinary excretion) for such measurements. Currently, no other clearance methods have proven suitable. Hepatic blood perfusion can be measured by dynamic PET using tracers for which the permeability of the plasma-hepatocyte membrane is high. Quantification of the hepatic blood perfusion requires measurements of the supply of tracer from an artery and the inaccessible portal vein, but methods have been developed that replace the invasive portal venous samples using tissue-derived information or a portal venous model. The methods have been developed and validated in pig studies and are promising for transfer to humans. It has been shown that the regional perfusion can be estimated using a short three-min dynamic PET scan employing the glucose analogue [¹⁸F]fluoro-2-deoxy-D-glucose (¹⁸FFDG) by compartmental model analysis. The hepatic blood perfusion can also be estimated using intravascular tracer ¹⁵O-carbon monoxide by microvascular models that account for the tracer concentration gradient along the liver sinusoids.

Measurement of hepatic blood perfusion and of the exchange of substances between blood and cells is a challenge. Long before the development of PET, multiple indicator data were analysed using models based on elaborate capillary theories. Today, PET is a unique modality that allows external quantitative measurements of the regional distribution of intravenously injected radiotracers and their metabolites in tissues, but dynamic PET data are analysed using less physiologically based schemes. The standard compartment model is an inlet equilibration model that does not naturally incorporate blood flow, and a single-uptake model that does not allow substances to re-enter the capillaries. These deficiencies lead to paradoxes when modelling fast blood-cell exchange. We have combined compartmental and capillary theory and developed microvascular models that account for blood flow and concentration gradients in capillaries. The microvascular models can be regarded as revisions of the input function which include more physiological realism and provide a superior description and interpretation of dynamic PET data when compared to the standard compartmental scheme.

PET is a powerful technique for measuring the regional metabolism of radiolabelled tracers in vivo, but the question remains how to convert the clearance of an analogue tracer measured by PET to the clearance of its mother substance, e.g., for [¹⁸F]fluoro-2-deoxy-D-glucose (¹⁸F-FDG) and glucose. The ratio between the extraction fraction of the tracer (E*) and that of the mother substance (E), as measured by arteriovenous concentrations across the organ, is often used as a proportionality factor for converting the PET-determined clearance of tracer to the clearance of the mother substance. However, this approach is founded in compartmental modelling which assumes a well-mixed concentration in the vascular compartment equal to the inlet (arterial) blood concentration. For the in vivo situation with perfused capillaries (for the liver, sinusoids), the removal of substrate from the blood by the cells creates concentration gradients in the capillary/sinusoid. We recently derived and validated that the lumped constant (Λ) equals ln(1–E*)/ln(1– E), being a correct in vivo proportionality factor in capillary beds for the removal of tracer and mother substance in the cells. This relationship is independent of the concentration of the mother substance, whereas E*/E may vary with the concentration of the mother substance.

The liver is a vital organ involved in the preservation of the balance among the metabolic functions of the whole-body. Liver metabolism is complex and most difficult to investigate or to monitor, and only indirect measurements have been available in human beings for a long time. This is due to the peculiar anatomical location of the liver, since circulating substrates reach the organ via the hepatic artery and the portal vein, the latter being not accessible for routine measurements. Recently, PET imaging has been introduced to characterize liver glucose and fatty acid metabolism. Mathematical modelling of liver metabolism by use of PET and [¹⁸F]fluoro-2-deoxy-D-glucose (¹⁸F-FDG) or [¹¹C]Palmitate (¹¹C-palmitate) has been recently validated, and the requirement of a dual (arterial and portal venous) input function has led to the development of approaches for its estimation. The application of the methodology to the understanding of human metabolic disorders (obesity, impaired glucose tolerance, diabetes) highlights a tight reciprocal regulation between glucose and fatty acid metabolism, suggesting that the liver is first engaged in counterbalancing the overflow of fatty acids deriving from a dysfunctional adipose tissue, by the activation and/or exhaustion of oxidation and lipid accumulation. These events may compromise hepatic insulin action and glucose metabolism. Once chronic hyperglycaemia ensues, the capability of the liver to utilize fatty acids during fasting and glucose during insulin stimulation may become further defective. This chapter describes the methodological advances, and the clinical findings so far achieved by PET imaging in the field of liver metabolism in metabolic diseases.

Bile is produced by the liver and stored in the gallbladder during fasting. With eating, bile is discharged into the small intestine. As a digestive secretion, bile acts as a detergent by forming mixed micelles with lipid nutrients, thereby enhancing their absorption. As an excretory secretion, it delivers lipid waste products to the intestine from which they are poorly absorbed. The main component of bile is conjugated bile acids, which are amphipathic steroids formed in the liver as the end-products of cholesterol metabolism. Bile acids are recycled between liver and small intestine (i.e. the enterohepatic circulation) with low plasma concentrations. In addition to providing the detergent property of bile and being a major driving force of bile flow, bile acids are signalling molecules that carry signals from the intestine to the liver. Homeostasis of bile acid synthesis and ileal conservation is essential not only to achieve the physiological functions of bile acids, but also to avoid pathological effects caused by their amphipathic property. Bile acid retention in the hepatocyte due to bile duct disease (e.g. PBC, PSC) leads to cell death. In terms of functional molecular imaging, dynamic planar cholescintigraphy is currently the modality of choice for the evaluation of the biliary system often with Tc-99m-mebrofenin as the radiotracer. Hepatic bile acid synthesis and handling, as well as intestinal malabsorption of bile acids, have been assessed by dynamic planar cholescintigraphy using the labelled bile acid analogue ⁷⁵SeHCAT. Positron emission tomography (PET) using radiolabelled bile acid analogues has, to date, only been applied in preliminary studies regarding biliary secretion.

Metabonomics is a new technology that provides rapid, accurate and precise analysis of a large number of metabolites in biological systems. Using multivariate analysis and pattern recognition, it provides a direct unbiased measure of biochemical consequences of pathology or drug treatment. The integration of the all ‘omics’ information (genomics, transcriptomics, proteomics and metabonomics) can provide knowledge for better understanding of biological systems and may lead to the generation of new mechanistic hypotheses to explain the development of pathologies. In this chapter, we will give an overview of the basic principles of this new technology and provide examples of applications in liver diseases.

Hepatocellular carcinoma (HCC) is the most common primary liver cancer. HCC typically develops in a cirrhotic liver and the overall survival is poor, primarily because the disease is often diagnosed at an advanced stage, when curable treatment is no longer an option. Ideally, HCC lesions should be detected when they are ≤ 2-3 cm in diameter. The diagnosis of HCC has improved with the introduction of 4-phase contrast enhanced CT and dynamic, contrast enhanced MRI. However, for small lesions, which are potentially curable, the detection rate remains poor. Molecular imaging of specific metabolic pathways may provide new opportunities for early diagnosis of HCC.

Cholangiocarcinoma (CC) is the second most common primary hepatic malignancy. Primary sclerosing cholangitis (PSC) predisposes for CC with a yearly incidence of 1.5%. Symptoms of CC are late and therefore the prognosis is poor. Ultrasonography (US), computer tomography (CT), magnetic resonance imaging (MRI), and cholangiographic techniques with cytology are the best imaging modalities, but these techniques are less useful for detection of tumour recurrence after surgery and metastases. Positron emission tomography (PET) using [¹⁸F]fluoro-2-deoxy-D-glucose (¹⁸F-FDG PET) has a diagnostic sensitivity for diagnosis of the primary tumour in CC complicating PSC ranging from 0-1.0 and a specificity of 0.67-1.0 in small studies of advanced PSC. The sensitivity and specificity of ¹⁸F-FDG PET for diagnosis of CC in patients with suspected bile duct cancer were 0.61-0.94 and 0.80- 1.0, respectively. With ¹⁸F-FDG PET/CT, sensitivity and specificity were 0.68-1.0 and 0.63-1.0, respectively, in this group of patients. Mass lesions and intrahepatic CC were more readily diagnosed than tumours with an infiltrating growth pattern and extrahepatic CC. Direct comparison of ¹⁸F-FDG PET, ¹⁸F-FDG PET/CT on 1 hand and CT on the other for diagnosis of the primary tumour showed no significant differences in sensitivity or specificity between the modalities. This also holds true for detection of regional lymph node metastases. However, ¹⁸F-FDG PET and ¹⁸F-FDG PET/CT performed better than CT for diagnosis of both recurrence of CC and distant metastases. The usefulness of ¹⁸FFDG PET/CT for screening and surveillance of PSC-patients for CC remains to be established.

Colorectal cancer is the third most common malignancy in both sexes with increasing incidence with age. In 2010, approximately 142,570 new cases of colorectal cancer (CRC) will result in it being the second leading cause of cancer death in the United States. Approximately 70-80% of patients with CRC are surgically treated with curative intent. The overall survival at 5 years is 67%. Approximately 25% of patients will have hepatic metastases at the time of diagnosis and an additional 25% will later develop hepatic metastases. About 14,000 (20%) of patients will die annually with metastases to the liver as the only site of disease spread. Selective hepatic resection is the only curative therapy for these patients, but it has a high mortality rate (2-7%). Therefore careful pre-resection screening is crucial to avoid futile partial hepatectomy. Imaging has a central role in accurate staging of CRC, via liver ultrasound, CT of the chest, abdomen and pelvis, MRI of the liver and whole-body 18FFDG PET/CT, with each imaging modality having a proper role. The single best imaging modality for detection of intra- and extra-hepatic metastases is ¹⁸F-FDG PET/CT. ¹⁸F-FDG PET/CT can also be used to evaluate early treatment response. ¹⁸F-FDG PET does have limitations including false-positives due to treatment induced inflammation and false negatives from poor ¹⁸F-FDG uptake in CRC with a high mucinous component or the transient effects of chemotherapy. Despite these limitations, 18F-FDG PET/CT is a valuable tool in the evaluation of liver metastases from colorectal cancer.

Secondary liver lesions are frequently encountered in neuroendocrine tumours (NET) of the gastro-entero-pancreatic tract. Although conventional morphologic imaging modalities and somatostatin receptor scintigraphy (SRS) have been extensively used in past decades, with the advent of PET/CT, new diagnostic procedures are currently available with the employment of different beta emitting tracers. In fact NET can be easily visualised on PET scans using an array of both metabolic and receptor-based tracers. 3,4-dihydroxy-6-[¹⁸F]fluoro-L-phenylalanine (¹⁸F-DOPA) and ⁶⁸Ga-DOTApeptides (DOTA-TOC, DOTA-NOC, DOTA-TATE) are very promising to image well-differentiated NET and were reported to be superior to other imaging modalities (CT, SRS). On the contrary, the role of ¹⁸F-FDG is limited in well-differentiated NET, due to their low glucose metabolism and growth rate, while it can still provide valuable information in less differentiated NET or in forms with lower expression of somatostatin receptors.

Brain metabolism is in principle similar to the biochemical processes which take place in all cells and organs. However, specialization of the different cell types in the brain, i.e. neurons or glial cells, adds a complexity which differs from other organs. In this chapter, a detailed outline of the metabolic processes providing the basis for brain energy and ammonia metabolism is provided. It also contains information regarding the effects on energy metabolism of a dysfunctional ammonia homeostasis such as that related to hepatic encephalopathy (HE). Moreover, methodological tools by which these metabolic processes may be studied experimentally such as PET as well as tissue and cell preparations are reviewed.

Hepatic encephalopathy shows very distinct clinical features combining cognitive dysfunction, alterations of consciousness and motor symptoms indicating an alteration of the extrapyramidal and cerebellar motor system and the pyramidal tract. Positron emission tomography studies of the cerebral glucose utilisation in patients with liver cirrhosis and various grades of hepatic encephalopathy show accordingly alterations in distinct brain areas – a decrease of the glucose utilisation especially in the frontal cortex and anterior cingulated gyrus, and an increase in the hippocampus, basal ganglia and cerebellum. These alterations in cerebral glucose utilisation correlate significantly with psychometric test results and brain metabolite alterations as measured by magnetic resonance spectroscopy.

Ammonia is thought to play a significant role in the development of hepatic encephalopathy (HE) in patients with liver failure. Some of the key questions have been whether there is an abnormal permeability surface-area product across the blood-brain barrier (PS_BBB) for ammonia in patients with cirrhosis and whether the cerebral metabolic rate of ammonia is altered. In this chapter, we discuss how functional molecular imaging using positron emission tomography (PET) can be utilized for in vivo studies of cerebral ammonia kinetics and what the studies so far have shown.

The diverse neurological manifestations of liver disease pose a diagnostic challenge for modern hepatologists with magnetic resonance (MR) techniques representing a highly profitable avenue to investigate structural and functional aspects of brain dysfunction. The commonest, but not sole neurological manifestation of liver disease, hepatic encephalopathy (HE), is a sequel of both acute and chronic liver failure, ranging from cognitive impairment, only detectable on psychometric evaluation through to confusion, coma and death from cerebral edema. While there is widespread acceptance of its importance, there is little consensus on how best to diagnose and monitor HE. Clinical descriptions, psychometric testing, electroencephalography (EEG) and lately, imaging techniques, such as magnetic resonance (MR) imaging, MR spectroscopy and positron emission tomography (PET) all have their proponents, but most putatively, diagnostic paradigms remain the preserve of research establishments with an interest in HE. Of note, modern clinical MRI scanners with multinuclear MR spectroscopy capabilities and brain mapping software can objectively demonstrate structural and functional cellular changes (such as brain size and astrocyte swelling) using volumetric MRI, magnetization transfer MRI, diffusion-weighted MRI, functional MRI with oxygenation measurements and in vivo ¹H and ³¹P MR spectroscopy, with the option of performing many sequences at a single session to maximize information gathering and cohort characterization. These techniques are also transferable to characterize non-HE cases of neurological dysfunction in liver disease, such as primary biliary cirrhosis, the neuropsychiatric sequela of non-cirrhotic chronic hepatitis C infection and metal deposition disorders, such as Wilson’s disease and hemochromatosis. This chapter describes the relative merits of these MR techniques and provides guidance on the directions for future research and translational into clinical practice.