Antioxidants for preventing gastrointestinal cancers
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To assess the beneficial and harmful effects of antioxidants in prevention of gastrointestinal cancers (esophageal, gastric, small intestinal, colorectal, pancreatic, liver, and biliary tract cancers, with antioxidants based on randomised trials.
Background
The recent progress in molecular biology has led to identification of lesions in genes and other cellular components involved in initiation and promotion of cancer. Because carcinogenesis is a multistage process with a latency of many years, there is a considerable opportunity for intervention. Use of agents that can inhibit, delay, or reverse carcinogenesis is called chemoprevention (Tamimi 2002). Many classes of agents including antioestrogens, anti‐inflammatories, and antioxidants have shown promise in chemoprevention (Tamimi 2002).
Although cancers of the gastrointestinal tract harbour many dissimilarities regarding etiological factors, growth patterns, histology, etc., they share three common traits. First, all cells of the gastrointestinal tract originate from the same cell line, the endoderm. Second, the gastrointestinal tract is exposed to the oral carcinogenetic factors. Third, our knowledge about development and prevention of these cancers is still limited. In this review we will consider the chemoprevention with antioxidants of the following gastrointestinal cancers: esophageal, gastric, small bowel (small intestinal), large bowel (colorectal), pancreatic, liver (hepatocellular carcinoma), and biliary tract cancers.
Oxidative stress, defined as a disturbance in the prooxidant‐antioxidant balance (Sies 1985) is considered as a causative factor in cancer (Ames 1995). Oxygen free radicals are extremely reactive chemical species that can damage DNA, proteins, and membranes of the cells. They can also modify DNA bases causing production of DNA adducts, which can initiate carcinogenic process by producing mispaired DNA sequences (Kasai 1984; Nestmann 1996; Sihvo 2002). In malignancies of the gastrointestinal tract increased levels of DNA adducts appear (Lee 1998; Wang 1998; Janne 2000; Sihvo 2002).
Antioxidants are compounds used by aerobic organisms for protection against oxidative stress induced by free radicals (Papas 1999; Tamimi 2002).The possibility that high intake of antioxidants may allow protection against cancer has drawn much attention in last decades (Diplock 1994; Poppel 1997). Epidemiological (Schrauzer 1977) and laboratory (Peto 1981) studies suggest a role of diet supplementation in cancer.
There are pieces of evidence that oxidative stress has a role in development of esophageal cancer through malignant transformation of Barrett's esophagus (Sihvo 2002). There is inverse association between fruit and vegetable consumption and the risk of esophageal and gastric cancer (Block 1992; Wang 1994; Steinmetz 1996). Some components of fruits and vegetables like vitamin C and E, carotenoids, dietary fiber, and flavonoids may be responsible for this protective effect. Possible mechanisms through which these components may act include antioxidant effects (vitamins C, E, carotenoids, and flavonoids), inhibition of nitrosamine formation (vitamins C and E), and dilution and binding of carcinogens (fiber) (Botterweck 1998; Botterweck 2000).
Small intestinal cancers are very rare and prevention with antioxidants has according to our information not been tested yet.
Colorectal cancer develops through a multistep process arising as the result of a series of molecular changes that transform normal colonic epithelial cells into colorectal cancer with an adenomatous polyp as an intermediate step in this process (Ponz de Leon 1997; Janne 2000). Adenomatous polyps can be found in 33 to 50 per cent of the general population aged 50 to 70 years (Williams 1982). Colorectal cancer prevention may therefore be directed at preventing the development of adenomatous polyps and/or their subsequent progression to colorectal cancer using a variety of oral agents mainly containing antioxidants (Janne 2000). It is postulated that people who consume a diet high in vegetables and fruits, rich in antioxidants, have a lower risk of colorectal cancer (Boyle 1985).
Possible etiologic factors for pancreatic cancer include chronic pancreatitis, alcohol consumption, cigarette smoking, diabetes, and other medical conditions (Haddock 1990). Chronic inflammation, resulting in chronic phagocytic activity, which is one of the major endogenous sources of free radicals, is associated with cancer of several organs (Collins 1987; Shimoda 1994; Holzinger 1999). Dietary components may also be protective in cases of pancreatic cancer (Rautalahti 1999).
Hepatocellular carcinoma is a malignancy with increased incidence over the last two decades of the 20th century (Akriviadis 1998) and its pathogenesis is still poorly understood. Cirrhosis has been recognized as a main risk factor for hepatocellular carcinoma regardless of the cause (alcohol abuse; hepatitis B virus; hepatitis C virus) (Edmondson 1956; Johnson 1987; Kew 1997) and has been considered a preneoplastic condition. Other risk factors are increasing age, chronic hepatitis C virus, hepatitis B virus infection (Yu 1997), and aflatoxins (Yu 1997; Akriviadis 1998). The latter are in some areas of East and South Africa considered to be responsible for up to a half of cases of hepatocellular carcinoma (Anthony 2001). Oxidative stress has been documented in patients with hepatitis, cirrhosis, and liver cancer (Shimoda 1994; Yamamoto 1998). In the liver it can modulate the apoptotic programme (a programmed cell death) by inhibiting it and promoting hepatocarcinogenesis (Patel 1998). Viral or chemical damage to the liver results in oxidative damage, producing free radicals that can act as tumour promoters, leading to hepatocellular carcinoma (Tabor 1999; MacDonald 2001). The liver is well endowed with antioxidant mechanisms to combat oxidative stress, including micronutrients, such as vitamin E and vitamin C, and some enzymes that metabolise reactive metabolites and reactive oxygen species (Kaplowitz 2000).
The role of antioxidants in preventing the biliary tract cancers based on recent evidences, is not sufficiently investigated. There are only few experimental studies dealing with this problem (Takeda 2002).
A normal healthy diet should provide sufficient antioxidants like vitamin A and carotenoids, vitamin C, vitamin E, and selenium, but inflammatory conditions, alcohol, and smoking may tip the balance in favour of oxidative stress and risk of cancerogenesis (Sihvo 2002).
Vitamin A is essential for growth, and since cancer is a disease involving disturbances in normal tissue growth and differentiation, it was one of the first vitamins, which role was evaluated in respect to cancerogenesis. Later studies indicated that protective effects were only observed for dietary vitamin A from plant sources (beta‐carotene) (Peto 1981; Ziegler 1989). Beta‐carotene belongs to the carotenoids, which are compounds classified as xanthophylls, carotenes, or lycopene.
Vitamin C has markedly antioxidative properties with possible cancer preventive potential (Hanck 1988) acting as a free radical scavenger, preventing carcinogenic nitrosamine formation. These protective mechanisms may explain the observation that consumption of vegetables and fruits is associated with a reduced risk of cancer.
Vitamin E is acting as a free radical scavenger to prevent lipid peroxidation of polyunsaturated fatty acids and block nitrosamine formation (Oshima 1982; Poppel 1997). Vitamin E supplementation can increase production of humoral antibodies and has been hypothesised to have antitumour proliferation capacities, possibly by modulating gene expression (Knekt 1994).
Selenium, a trace element, is also important for the antioxidant defences of the body as integral component of metalloprotein enzymes. It is a component of selenoproteins, which have important enzymatic functions (Hughes 2000; Rayman 2000). Deficiency of selenium is accompanied by loss of immunocompetence. It is documented that there is an inverse relation between selenium intake and cancer mortality (Schrauzer 1977).
However, the evidence on whether antioxidants are effective in decreasing the incidence of gastrointestinal cancers are contradictory (Nomura 1987; Dawsey 1994; Greenberg 1994; Wang 1994; Yu 1997; Yu 1999). Some trials having lung cancer as main outcome raise the possibility that these supplements may have harmful effects and even co‐carcinogenic properties. Beta‐carotene supplementation was associated with a higher risk of lung cancer, cardiovascular disease, and total mortality when tested against placebo in randomised trials (Anonymous 1994, Omenn 1996, Omenn 1996a). We have been unable to identify any meta‐analyses or systematic reviews on antioxidants for gastrointestinal cancers.
Objectives
To assess the beneficial and harmful effects of antioxidants in prevention of gastrointestinal cancers (esophageal, gastric, small intestinal, colorectal, pancreatic, liver, and biliary tract cancers, with antioxidants based on randomised trials.
Methods
Criteria for considering studies for this review
Types of studies
Randomised trials only, irrespective of blinding, publication status, publication year, or language.
Types of participants
-
participants from the general population at any age, sex, or ethnic origin
-
participants with higher risk of development of gastrointestinal cancers
-
participants coming from other patient groups, primarily with non‐gastrointestinal diseases.
Types of interventions
Antioxidants (vitamin A and carotenoids, vitamin C, vitamin E, and selenium, and other administered antioxidants that we might identify) at any dose, duration, and route of administration, administrated separately or in combination versus placebo or no intervention. Studies concerning antioxidants in prevention of other organ system disease (cardiovascular, urinary tract, etc.) will be considered if data on the incidence of gastrointestinal cancers during the trial can be obtained.
We will analyse the antioxidants individually as well as combined.
Collateral interventions will be allowed if used in both intervention arms of the trial.
Types of outcome measures
Primary outcome measures
(1) Number of patients developing gastrointestinal cancers.
We will determine whether supplementation with antioxidants, administrated separately or in combinations, can reduce the incidences of individual gastrointestinal cancers (esophageal, gastric, small intestinal, colorectal, pancreatic, liver, and biliary tract cancers) and all gastrointestinal cancers combined.
(2) Overall mortality.
We will determine whether supplementation with antioxidants could reduce overall mortality in trials including participants from the general population as well as populations with high risk of developing gastrointestinal cancers. However, due to the risk of beneficial or harmful interaction between antioxidants and mortality in populations with other diseases, these studies will only be included in exploratory analyses.
Secondary outcome measures
(1) Any adverse events as reported in trials.
Depending on availability of data, we will attempt to classify the adverse events as serious or non‐serious. Serious adverse events will be defined as any untoward medical occurrence that was life‐threatening, resulted in death or persistent or significant disability, or any medical event, which may have jeopardised the patient or required intervention to prevent it (CFR & ICH 1997). All other adverse events will be considered non‐serious.
(2) Quality of life measures.
(3) Health economics.
Search methods for identification of studies
We will ask the trials search coordinators of The Cochrane Colorectal Cancer Group, The Cochrane Inflammatory Bowel Disease Group, The Cochrane Hepato‐Biliary Group, The Cochrane Upper Gastrointestinal and Pancreatic Diseases Group, as well as of The Cochrane Cancer Network to provide us with searches performed on their respective trials registers on antioxidants and prevention of esophageal, gastric, small intestinal, colorectal, pancreatic, liver, and biliary tract cancers.
Further, we will conduct electronic searches on The Cochrane Central Register of Controlled Trials on The Cochrane Library (Issue 1, 2003) as well as PubMed/MEDLINE (1966 to February 2003), EMBASE (Excerpta Medica Database) (1985 to February 2003), and LILACS (1982 to February 2003). We will also search the Science Citation Index Expanded (SCI‐EXPANDED) ‐ 1945 to February 2003 (Web of Science at http://isi2.isiknowledge.com/portal.cgi/wos). The initial search strategies that we will test and will try to improve during the performance of the searches are given in 'Table 1' (see 'Additional tables' ).
| Database | Search performed | Search strategy |
| The Cochrane Central Register of Controlled Trials on the Cochrane Library | DIGESTIVE SYSTEM NEOPLASMS explode all trees (MeSH) ‐ 4051 hits, | |
| MEDLINE | #1. (antioxidant* | |
| EMBASE | #1. (antioxidant* | |
| LILACS | We will combine each line from group A with each line from group B and with the word 'random*'. | |
| The Web of Science | We will perform a search similar on MEDLINE and EMBASE. | |
| The Chinese Biomedical Database | We will present the search strategy with the publication of the review. | |
| The Controlled Trials Registers of the four Cochrane gastrointestinal groups | We will present the search terms used with the publication of the review. | |
In addition, we will ask The Chinese Cochrane Centre to perform a search for us on the Chinese Biomedical Database.
We will scan reference lists from review articles retrieved from the searches above in order to identify additional trials.
We will contact manufacturers of antioxidants to ask for unpublished randomised trials.
Data collection and analysis
Inclusion criteria application
We will retrieve the identified material for assessment. GB and DN will independently apply the inclusion criteria to all potential studies. We will perform this without blinding. When a discrepancy occurs in the trial selection, GB and DN will ask the third reviewer (RS) for an opinion in order to reach consensus.
Data extraction
‐ Patient characteristics, diagnosis, and treatments
We will record the following data from the individual randomised trials: first author; country of origin or region of study; number of participants, characteristics of patients studied: age range ‐ mean or median, sex ratio, years followed; participation rate; drop‐out rate; type and dose of the antioxidant intervention (route of administration, formulation, frequency, and duration of dosing); type of intervention in the control group as well as any co‐interventions. In patients with:
Esophageal cancer ‐ incidence of different histological types (squamous cell carcinoma, adenocarcinoma of the esophagus or esophagogastric junction), or overall incidence, if such data are not available. Presence or absence of conditions that appear to increase the risk of squamous cell carcinoma (long standing achalasia or caustic strictures, Patterson‐Kelly syndrome) and adenocarcinoma (Barrett's esophagus) will also be recorded as well as risk factors including smoking and gastro‐esophageal reflux.
Gastric cancer ‐ distribution of cancer within the stomach based on endoscopic findings and histologic type of cancer. Incidence of premalignant conditions (chronic atrophic gastritis, intestinal metaplasia, gastric dysplasia, gastric polyps) and risk factors like smoking and Helicobacter pylori infection will also be recorded.
Small intestinal cancer ‐ distribution of cancer within the small intestine (duodenum, jejunum, or ileum) and histologic type.
Colorectal cancer ‐ distribution of cancer within the colon and rectum based on endoscopic findings, (incidence of colon and rectal cancers separately as well together; histologic type. Existing adenomas and their size before and after treatment and premalignant conditions like familiar adenomatous polyposis or inflammatory bowel diseases will also be recorded.
Pancreatic cancer ‐ distribution of cancer within the pancreas (head, body, or tail), tumour size, and histologic type. Presence or absence of premalignant conditions like chronic pancreatitis and risk factors like smoking and alcohol consumption will also be recorded.
Hepatocellular carcinoma ‐ form of liver disease according to the etiology (hepatitis B; hepatitis C; alcoholism, other); duration of liver disease; severity of the disease at entry; alcohol consumption at entry and during the follow‐up. We will also register histologic type; development of clinical symptoms and complications; adverse events during follow‐up.
Biliary tract cancers ‐ distribution of cancer within the biliary tract (gallbladder, extrahepatic biliary tree, or ampulla of Vater) and histologic type.
‐ Trial characteristics
We will record the date, location, sponsor of the trial (known or unknown and type of sponsor) as well as publication status.
‐ Methodological quality
The methodological quality, defined as the confidence that the design and report will restrict bias in the intervention comparison (Moher 1998), will be evaluated independently and unblinded by two reviewers (BG and DN). According to the empirical evidence (Schulz 1995; Jadad 1996; Moher 1998; Jüni 2001; Kjaergard 2001), we will assess the methodological quality as described by Kjaergard et al. (Kjaergard 2001):
(1) generation of the allocation sequence: adequate (computer generated random numbers, table of random numbers, or similar) or inadequate (other methods or not described);
(2) allocation concealment: adequate (central independent unit, sealed envelopes, or similar) or inadequate (not described or open table of random numbers or similar);
(3) double blinding: adequate (identical placebo tablets or similar) or inadequate (not performed or tablets versus injections or similar); and
(4) follow‐up: adequate (number and reasons for dropouts and withdrawals described) or inadequate (number or reasons for dropouts and withdrawals not described).
We will also report on whether the investigators have performed a sample size calculation and used an intention‐to‐treat analysis (Gluud 2001).
We will use the results of the methodological quality for sensitivity analysis and not as exclusion criteria. We will solve disagreements by discussion.
Analyses and presentation
We will perform the meta‐analyses according to the recommendations of The Cochrane Collaboration (Clarke 2002). We will list identified trials and record their fulfillment of the inclusion criteria. We will also list excluded trials with the reason for exclusion.
We will use the software package (RevMan 4.1) provided by The Cochrane Collaboration. For dichotomous variables, we will calculate the relative risks with 95% confidence interval. Based on the presence or absence of significant heterogeneity (P < 0.1), we will use a random effects model (DerSimonian 1986) or fixed effect model (DeMets 1987). The analyses will include all patients irrespective of compliance or follow‐up, according to the 'intention‐to‐treat' principle, and using the last reported observed response ('carry forward'). In addition, we will perform 'a worst case ‐ best case scenario' analysis, in which we will consider participants with missing data as treatment failures in the experimental arm and patients with missing data in the control arm as treatment successes.
We will also use funnel plot asymmetry to assess the existence of publication bias and other biases (Egger 1997).
We will analyse all data as described below:
-
participants from the general population at any age, sex, or ethnic origin
-
participants with higher risk of development of gastrointestinal cancers
-
participants coming from other patient groups, primarily with non‐gastrointestinal diseases.
If we can identify enough trials we are going to perform sensitivity analysis on:
-
methodological quality of the trials
-
duration of the treatment and dose of the supplement.
The protocol for this systematic review is a collaborative endevour between The Cochrane Upper Gastrointestinal and Pancreatic Diseases Group, The Cochrane Hepato‐Biliary Group, The Cochrane Inflammatory Bowel Disease Group, and The Cochrane Colorectal Cancer Group; it has been primarily designed by The Cochrane Hepato‐Biliary Group, and then significantly broadened in scope after implementing the suggestions received by the coordinating editors of the above mentioned groups.
| Database | Search performed | Search strategy |
| The Cochrane Central Register of Controlled Trials on the Cochrane Library | DIGESTIVE SYSTEM NEOPLASMS explode all trees (MeSH) ‐ 4051 hits, | |
| MEDLINE | #1. (antioxidant* | |
| EMBASE | #1. (antioxidant* | |
| LILACS | We will combine each line from group A with each line from group B and with the word 'random*'. | |
| The Web of Science | We will perform a search similar on MEDLINE and EMBASE. | |
| The Chinese Biomedical Database | We will present the search strategy with the publication of the review. | |
| The Controlled Trials Registers of the four Cochrane gastrointestinal groups | We will present the search terms used with the publication of the review. | |
