An important advance in cancer treatment has been made in recent years with the finding that adjuvant therapy can significantly improve the survival of patients with colorectal cancer. In patients with resected lymph node-positive colon carcinomas (TNM stage 3), adjuvant 5-fluorouracil and levamisole produced an unequivocal survival advantage that established this combination as the standard of clinical practice. Given that biochemical modulation of fluorouracil by leucovorin can increase response rates in advanced disease, this combination is undergoing evaluation as an adjuvant treatment. Preliminary results indicate that 5-fluorouracil and leucovorin are effective in reducing disease relapse; however, the effect of this regimen on patient survival rates awaits extended follow-up. In the treatment of stages 2 and 3 rectal cancer, significant reductions in local recurrence and death rates have been achieved with the combination of 5-fluorouracil and radiation therapy. Immunologic approaches and newer chemotherapeutic agents may further improve patient outcome and are under investigation, as are efforts to reduce the toxic effects of cancer chemotherapy. Increased understanding of the biology of these diseases is likely to yield prognostic markers capable of identifying subgroups of earlier stage patients at high risk of disease relapse who may also benefit from adjuvant therapy.

Determining the role of eicosanoids in gastrointestinal physiology and pathophysiology has been an active area of investigation over the past 20 years. The landmark discovery of prostaglandin endoperoxide synthase and other enzymes involved in the production of arachidonic acid products (lipoxygenases and epoxygenases) ushered in a new era of research. The goal of this review is to distill a large body of work pertaining to studies of eicosanoids in the gastrointestinal tract. This review has been organized according both to functional (secretion and motility) and disease-related (inflammation, mucosal injury, and neoplasia) effects. The aim of this article is to present a clear summary of this area of gastroenterology so that future research can be directed in a logical and productive manner.

The role of acid and duodenogastric reflux (DGR) in the development of esophageal mucosal injury has been extensively investigated using both animal and human models. In this report, clinical and experimental data are reviewed. The mechanisms by which gastric and duodenal contents produce esophageal mucosal injury are also discussed. Acid and pepsin are unquestionably important in causing mucosal damage at low pH values in both animal and human models. Animal models suggest synergistic damaging potential for conjugated bile acids and HCI as well as that of unconjugated bile acids and trypsin in more neutral pH values. Human evidence for the involvement of bile and its constituents has been controversial; however, the advent of better technology to detect DGR is beginning to clarify the role of these constituents. The contribution of each methodology in clarifying the extent of involvement of DGR in esophageal mucosal injury is reviewed. Despite some conflicting results, preliminary human studies support the results from the animal data suggesting synergistic damaging effects for both bile and acid in esophageal mucosal injury. The implication of these studies in treating gastroesophageal reflux disease are discussed.

Recent research shows that each word in the definition of coeliac disease, permanent gluten sensitive enteropathy, must now be reviewed, revised or reinterpreted. Permanent--but there are now well-documented cases of acquired disease, and perhaps also partial recovery of gut gluten tolerance. Enteropathy--gluten sensitivity is expressed in a spectrum, with a mild form seen as normal architecture with high count of intraepithelial lymphocytes. Gluten--the provoking agent--Investigators are intensively working to identify the precise toxic sequence, and to establish how this will link in with new genetic information. Mechanism of sensitivity? or hypersensitivity?--Critical to this is new knowledge on the modulation and regulation of immunity to intestinal antigens, including gliadin. A hypothesis is presented, as to the pathogenesis of gluten-sensitive enteropathy, which combines concepts of oral tolerance and of the regulation of expression of delayed type hypersensitivity reactions in the gut mucosa.

Dermatitis herpetiformis (DH) is a relatively rare skin disorder with an estimated incidence of 1:10,000 in the UK. It is characterized by urticarial plaques and blisters on the elbows, buttocks, and knees, although other sites may also be involved. The eruption tends to be persistent: only 10-15% of patients have spontaneous remission over a 25-year study period. The disease is characterized by the presence of IgA deposits in the upper dermis of uninvolved skin and the diagnosis should not be made in the absence of these deposits. Two-thirds of patients have a small intestinal enteropathy with villous atrophy as seen in coeliac disease (CD). However, the remaining third also show evidence of a gluten sensitivity in the intestine, as judged by increased lymphocytic infiltration of the epithelium. Villous atrophy also ensues after gluten challenge in those patients with previous normal villous architecture. The initial treatment of the rash is with one of the following three drugs, dapsone, sulphapyridine or sulphamethoxypyridazine. However, the rash also clears with gluten withdrawal. It must be stressed, however, that the average time to achieve significant reduction in drug requirements is 6 months and it can be over 2 years before drugs are no longer required. On re-introduction of gluten the eruption recurs. Patients with DH have a high incidence of auto-immune disorders, thyroid disease, pernicious anaemia, and insulin-dependent diabetes, and should be screened for those diseases on a yearly basis. As with coeliac disease there is also an increased incidence of lymphoma and a gluten-free diet appears to protect patients from this complication. The mechanism by which gluten causes the skin lesions has still to be elucidated, but current investigations implicate lymphocytes and cytokines in the pathogenesis. The original hypothesis of an antigen-antibody reaction in the skin with complement activation causing the skin lesions, may not be correct.

Neoplasms constitute the major complication of coeliac disease, and high-grade T-cell lymphoma of the small intestine (enteropathy-associated T-cell lymphoma) is the most common neoplasm in this category. HLA genotyping indicates that in patients with enteropathy-associated T-cell lymphoma have the coeliac disease associated DQA1*0501, DQB1*0201 phenotype, although additional HLA-DR/DQ alleles may represent risk factors for lymphoma development. Molecular biological and immunohistochemical studies have shown that the intestinal mucosa distant from the tumour contains clonal populations of small T cells, often of the same clone as the high-grade T-cell lymphoma. These findings suggest that enteropathy-associated T-cell lymphoma arises in the setting of coeliac disease and evolves from reactive intraepithelial lymphocytes through a low-grade lymphocytic neoplasm to a high-grade tumour, which is usually the cause of the presenting symptoms. Most cases of chronic ulcerative enteropathy (ulcerative jejunitis) are probably part of the same disease process. If the ulceration occurs at a time when the neoplastic T-cells are of a low grade, morphological recognition of tumour cells in the ulcers may be impossible. Carcinoma of the pharynx and oesophagus, and adenocarcinoma of the small intestine, are increased in frequency in patients with coeliac disease. The increased risk of carcinoma of the oesophagus may be related to vitamin A deficiency. A number of reports have indicated an increased prevalence of various types of chronic hepatitis in patients with coeliac disease, but no coherent view of the cause of this association has emerged. Similarly, patients with coeliac disease have been reported to have various forms of fibrosing lung disease of uncertain causation. In recent years, there have been several reports, mainly from Italy, of a syndrome of epilepsy and bilateral brain calcification occurring in coeliac patients. The pathogenesis of this condition is not known and its prevalence in other communities is uncertain. Splenic atrophy occurs frequently in patients with coeliac disease and is related to the severity of the disease and degree of dietary control. Splenic atrophy predisposes to infection with capsulated bacteria, although mortality studies indicate that infection with these organisms is not a major cause of death in patients with coeliac disease.

Coeliac disease is a chronic disease characterized by small bowel villous atrophy which impairs nutrient absorption and improves on withdrawal of wheat gliadins and barley, rye and oat prolamins from the diet. Knowledge of the adult form of coeliac disease has greatly improved in recent years. Although this knowledge is not yet sufficiently widespread among referring clinicians, it has, over the past few years, allowed an increasing number of patients to be diagnosed with subclinical forms characterized by minor, transient or apparently unrelated symptoms. As a consequence, our views on the clinical and epidemiological aspects of this condition, the prevalence of which in the general population is believed to be close to 1 in 300, have changed and are still changing. Since it has been demonstrated that a strict gluten-free diet is protective against the complications of adult coeliac disease, it is important that even subclinical and silent forms are diagnosed and treated as early as possible. Non-invasive screening tests, such as anti-gliadin and anti-endomysium antibody estimation, should therefore be used systematically in groups considered to be at risk of coeliac disease. These include first-degree relatives of coeliac patients and patients with insulin-dependent diabetes mellitus, iron-deficiency anaemia, epilepsy with cerebral calcification, recurrent aphthous stomatitis and dental enamel hypoplasia. Other conditions will probably be identified in the near future.

Coeliac disease usually presents in infancy or early childhood with diarrhoea, vomiting and interference with weight gain and growth. Withdrawal of dietary gluten is followed by resolution of the symptoms and signs and restoration of normal weight gain and growth; the characteristic subtotal villous atrophy of the jejunal mucosa also recovers. Later re-introduction of dietary gluten will lead to a return of the jejunal mucosal abnormality in the majority and to clinical relapse in many but not all. The severity and timing of both are variable and 5% of children initially considered on clinical, biopsy and gluten response evidence to have coeliac disease appear to develop permanent tolerance to gluten, although mucosal relapse may occur years after the re-introduction of dietary gluten in a minority, emphasizing the need for long-term follow-up. Although a diagnostic and subsequent follow-up jejunal biopsy are necessary to confirm the diagnosis, anti-gliadin IgA and IgG, anti-reticulum and anti-endomysium antibodies are now almost totally reliable in identifying children who have coeliac disease and are valuable in monitoring the adequacy of gluten withdrawal. Dietary compliance is frequently poor and regular supervision by a paediatric dietitian is needed; indeed, lifelong supervision to ensure gluten withdrawal is essential to reduce the chance of developing later gastrointestinal malignancy.

Gluten sensitivity is associated with a spectrum of mucosal lesions, arbitrarily termed pre-infiltrative, infiltrative-hyperplastic, flat-destructive and atrophic-hypoplastic. Histologically and immunohistologically these lesions are all compatible with T-cell-driven events operative at a local mucosal level. They are classifiable either in terms of antibody titres (pre-infiltrative) (see Chapter 10) or by the characteristic disposition of IELs throughout the surface and crypt epithelium. From in-vivo challenges, it has been demonstrated: (i) that all these lesions comprise a dynamically interrelated series of events, culminating in the severe flat-destructive lesion; and (ii) that gluten evokes a dose-responsive infiltration of IELs (CD3+ CD8+ and TCR alpha beta + or gamma delta +) into the epithelium. Apart from that, little is known of the functions of IELs; it is possible they may have little to do with the evolving mucosal pathology of gluten sensitivity. Increasing work seems to support a view, proposed from this laboratory over 10 years ago, that the immune-mediated responses in jejunal tissue in gluten sensitivity arise in the lamina propria, in association with DR+ macrophages and an abundance of CD4(+)-activated lymphocytes. Many other inflammatory consequences flow from these interactions, involving activation of mast cells, eosinophils and neutrophils, elaboration of cytokines and other products of inflammation, and increased hyperpermeability of the microvasculature with upregulation of adhesion molecules. The result is a doubling of lamina propria volumes in the severe flat lesion. Evidence is also given to show that measurable changes in lamina propria inflammation occur with the infiltrative-hyperplastic lesion. Symptomatology is not related to the degree of proximal mucosal pathology, but to the extent of the mucosal lesion. Data, although scanty, suggests that lesional pathology involves only 30-50% of the entire small bowel mucosa. Thus, most patients, irrespective of proximal mucosal damage, have latent (or asymptomatic) gluten sensitivity. Symptom development requires additional environmental triggers, of which infection is a major contributor. It should also be noted that, while these various environmental triggers may precipitate symptomatology, they do not advance the severity of the mucosal lesion.

Increasing evidence points to a direct role for T cells in the mediation of the coeliac intestinal lesion. There is good evidence for increased local T-cell reactivity, manifest as increased in T-cell activation in the lamina propria and T-cell proliferation in the epithelial compartment. A likely scenario is that gluten elicits antigen-specific responses by lamina propria T helper cells, probably of the Th1 (inflammatory-mediator) subtype, leading to secretion of pro-inflammatory cytokines. Such cytokines may have direct effects on intestinal enterocytes, as well as mediating indirect effects by upregulation of MHC antigens and by enhancing the activity of cytolytic T cells. Although gluten-specific IEL responses have not been demonstrated by intraepithelial T lymphocytes (IELs), increasing evidence suggests that IELs can act as cytolytic effector cells and hence are likely to exert enteropathic effects under the influence of pro-inflammatory cytokines.

IgA is transported into intestinal secretions to perform exclusion of luminal antigens. The prerequisites are antigen sampling by the Peyer's patch M cells, antigen processing by antigen-presenting cells, and presentation of antigenic peptides by HLA class II molecules to immunocompetent T-cells. The basis for intestinal immunity is the maturation cycle of specifically primed T and B cells from the gut-associated lymphoid tissue via mesenteric lymph nodes and peripheral blood back to the intestinal lamina propria. In coeliac disease, patients are sensitized against gluten and serum gliadin antibodies are often detected. Gliadin antibodies are also found in other gastrointestinal diseases, other disorders and in healthy individuals not carrying the coeliac disease-specific DQA/DQB alleles. On the other hand, serum reticulin and endomysium autoantibodies are both sensitive and highly disease-specific. Positivity in patients with normal jejunal morphology indicates latency of coeliac disease. These tissue autoantibodies are directed against fibroblast-derived extracellular matrix proteins. The immune system is involved in the amplification and perpetuation of the abnormalities of the intestinal mucosa in coeliac disease. The role of antibody in the pathogenesis remains unknown. The author hypothesizes gluten-triggered autoimmune mechanism to be operative.

Work continues to progress in the unravelling of the molecular interactions involved in the pathogenesis of coeliac disease. The immunogenetics of the disease implicate certain HLA DQ alleles as necessary for subsequent disease development. These HLA molecules have been shown to be necessary in the binding and presentation of gliadin peptides to antigen-specific T cells. Current work is examining the precise HLA-antigen interaction that may lead to the development of antigen-blocking agents. The isolation of antigen-specific T cells has led to the confirmation of a toxic T-cell epitope of the gliadin protein (residues 31-49) and it would appear likely that additional toxic epitopes may be similarly characterized in the near future. No common TCR motifs have so far been detected, although these may become apparent as this work progresses. The gliadin peptide sequence, residues 31-49, has now been demonstrated to be toxic in vivo. Additional toxic T-cell epitopes may also be present within gliadins, but this identification of a toxic gliadin sequence for the first time raises the possibility of future manipulation of the wheat genome (and other toxic cereals) that could lead to the development of new graminae cereals with the properties of wheat, but which do not induce toxicity in patients with coeliac disease.

In recent years, remarkable progress has been made in the elucidation of cereal protein structure and its relation to coeliac toxicity. Gluten proteins of wheat can be classified according to their primary structure into high-, medium- and low-molecular-weight (HMW, MMW, LMW) groups. Each of these groups contains two or three different protein types having partly homologous, partly unique, structural elements: chi- and gamma-type HMW subunits of glutenin (HMW group), omega 5 and omega 1,2-type gliadins (MMW group) and alpha-type gliadins, gamma-type gliadins and LMW subunits of glutenin (LMW group). Numerous proteins from the same type do exist with only a few modifications of the amino-acid sequence. The structure of the HMW and LMW group proteins can be divided into three and five domains, respectively. Most typical for each type and unique for cereals are the glutamine- and proline-rich domains containing repetitive sequences (HMW group: domain B; LMW group: domain I). omega-type gliadins consist almost entirely of repetitive sequences. Rye and barley, closely related to wheat, have protein types homologous to those of wheat. Early investigations showed that wheat gluten and, in particular, the alcohol-soluble gliadin fraction contained the factor toxic for coeliac patients. Equivalent protein fractions of rye, barley and probably oats were also considered to be toxic. The effects of toxic proteins were not destroyed by digestion with pepsin, trypsin and pancreatin. In-vivo (instillation) testing established the toxicity of alpha-type gliadins, and in-vitro (organ culture) testing of gliadin peptides demonstrated that the N-terminal region (domain I) of alpha-type gliadins is involved in activating coeliac disease. The longest sequences common for toxic peptides were found to be -Pro-Ser-Gln-Gln- and -Gln-Gln-Gln-Pro-. Various in-vitro tests and two in-vivo studies on synthetic peptides support the importance of one or both of these sequences. They do not occur in non-toxic food proteins and are characterized by their ability to form a beta-turn conformation. Although these sequences are probably not sufficient for toxicity in themselves, and other amino-acid residues are additionally required, they could serve as the starting point for further investigation.

The noninvasive assessment of intestinal permeability in humans has a 20-year history. Because the tests are increasingly used in clinical practice and research and because there is much controversy, we reviewed the literature and outlined the potential and possible shortcomings of these procedures. Data was obtained from personal files and from a systemic search through MEDLINE and EMBASE. The principle of the differential urinary excretion of orally administered test markers is explained with reference to the desired physicochemical properties of the markers and how the principle can be exploited to allow assessment of various other gastrointestinal functions. The use of intestinal permeability tests for diagnostic screen for small bowel disease and assessment of responses to treatment, the pathogenesis of disease, normal intestinal physiology, and the effect of drugs and toxins on the intestine is described and reviewed. The controversy surrounding the anatomic location of the permeation pathways that the markers use is highlighted. Noninvasive tests of intestinal permeability have fulfilled early promises of usefulness in clinical practice and research. There is now a need for integrated research into the basic mechanisms of regulatory control of the intestinal barrier function.