Scolaris Content Display Scolaris Content Display

Nonsteroidal anti‐inflammatory drugs (NSAID) and aspirin for preventing colorectal adenomas and carcinomas in general population

Esta versión no es la más reciente

Contraer todo Desplegar todo

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To determine if administration of nonsteroidal anti‐inflammatory drugs (NSAID) and/or aspirin is effective as chemoprevention for colorectal adenomas and carcinomas in general population compared to currently available other agents or no intervention and to assess adverse effects associated with the intervention. 

The following hypotheses will be tested:

1. NSAID and/or aspirin is better than no intervention

2. NSAID and/or aspirin is better than a placebo/sham intervention;

3. Aspirin with combination of resistant starch is better than no intervention;

4. One type of NSAID is better than another type;

5. Increased dose of NSAID and/or aspirin is better than another dose.

Background

Colorectal cancer (CRC) is a significant health burden globally and responsible for more than 600,000 deaths each year worldwide (Jemal 2011). It is also a major cause of morbidity, affecting quality of life either due to the disease itself or its treatment.

The vast majority of CRC is thought to arise from the malignant transformation of benign adenomatous polyps, through a multi‐step adenoma‐carcinoma sequence (Muto 1975), a process that may take 10 years or more before malignancy develops (Stryker 1987). Population CRC screening has been successfully introduced in the UK to detect and remove pre‐malignant polyps, thus preventing cancer development. However utilisation of CRC screening is incomplete and cancer is frequently detected at index examination (Atkin 2002).

Some patient groups have an increased risk of CRC and who require more intensive screening/surveillance. Of these, perhaps the most common group is those who have had a past history of colorectal adenomas or CRC. Less commonly there are those who are an increased risk of CRC due to genetic factors, where there is a wide spectrum of risk. It is possible to divide this population into three broad categories of risk for colorectal cancer: low, moderate and high risk (Cairns 2010). The high risk group comprises those with or at risk of a defined cancer predisposition syndrome (i.e. Lynch syndrome, familial colorectal cancer type X and the various polyposis syndromes).  These groups have very different screening needs due to their increased CRC risk and/or accelerated adenoma‐carcinoma pathway.

The concept of chemopreventive agents to reduce adenoma development or prevent malignant transformation would make an attractive adjunct to current screening programmes, particularly for those in with an increased risk of developing CRC.

We will conduct a systematic review of randomised controlled trail data to assess the role of NSAIDS and aspirin as a chemopreventive agent. Because of the problems identified above, the review will be subdivided as follows:

 

  1. NSAIDs and aspirin for the prevention of colorectal adenomas and carcinomas in general population (current review).

  2. NSAIDs and aspirin for the prevention of colorectal adenomas and carcinomas in patient with previous adenomas and/or genetic disposition (Cochrane Protocol 164).

  3. NSAIDs and aspirin for the prevention of recurrence and or metachronous colorectal carcinomas in patients previously treated for colorectal cancer. (Cochrane Protocol 184)

Description of the condition

Colorectal cancer (CRC) is a significant health burden globally, being the third most common malignancy in the UK and responsible for more than 600,000 deaths each year worldwide.

Description of the intervention

NSAID and aspirin are tablets that can be taken per os.

How the intervention might work

The exact mechanism of potential chemoprevention from NSAIDs and aspirin is unclear but is likely in part to be due to inhibition of cyclooxygenase 2, which is overexpressed in CRC and colorectal adenomas and which has a role in the regulation of angiogenesis, cell migration and attachment, and apoptosis.

Why it is important to do this review

There are numerous observational data to support nonsteroidal anti‐inflammatory drugs, including aspirin, having a chemopreventive effect.

However, it is difficult to make firm recommendation on the basis of observational studies, as the data are potentially biased due to the lack of randomisation. Furthermore when assessing studies for chemoprevention what should be the primary outcome measure? Adenoma development or recurrence is used as the primary outcome in most studies, but perhaps a more meaningful and hard primary outcome measure would be incidence of CRC. Finally as described above certain higher risk groups may derive particular benefit from a chemopreventive agent, for whom any risk/benefit analysis of a chemopreventive agent will be different compared with the general population.

Objectives

To determine if administration of nonsteroidal anti‐inflammatory drugs (NSAID) and/or aspirin is effective as chemoprevention for colorectal adenomas and carcinomas in general population compared to currently available other agents or no intervention and to assess adverse effects associated with the intervention. 

The following hypotheses will be tested:

1. NSAID and/or aspirin is better than no intervention

2. NSAID and/or aspirin is better than a placebo/sham intervention;

3. Aspirin with combination of resistant starch is better than no intervention;

4. One type of NSAID is better than another type;

5. Increased dose of NSAID and/or aspirin is better than another dose.

Methods

Criteria for considering studies for this review

Types of studies

RCTs and quasi‐randomised RCTs only will be included for the efficacy review.

All randomised or quasi‐randomised controlled trials comparing use of nonsteroidal anti‐inflammatory drugs and/or aspirin for preventing colorectal adenomas and carcinomas in general population with any alternative treatments or placebo will be reviewed to evaluate the chemoprevention effects.

Types of participants

This review will include studies on both adults and children,

The expected participants are:

‐People with no previous colorectal polyps

Types of interventions

1. Nonsteroidal anti‐inflammatory drugs (acetic acid derivatives, enolic acid derivatives, fenamic acid derivatives, p‐amino phenol derivatives, propionic acid derivatives,  selective COX‐2 inhibitors, sulphonanilides)

2. Aspirin (including lysine acetylsalicylate)

3. NSAID or Aspirin in combination with resistant starch, folic acid supplement, difluoromethylornithine, inulin/VSL#3

Types of outcome measures

The considered outcome measures are as followings:

Primary outcomes

  • Number of adenomas

  • Rate of new adenomas

  • Occurrence of carcinoma

Secondary outcomes

  • Polyp burden during the study period (total number, size of the largest polyp at the end of intervention, number of advanced adenoma, regression)

  • Drug adverse effects (cardiovascular toxicity, gastrointestinal toxicity, ototoxicity, thrombotic adverse event, interaction with chemotherapy)

Search methods for identification of studies

See: Cochrane Colorectal Group Methods used in reviews .

Electronic searches

Relevant trials will be identified from the Cochrane Register of controlled trials. The register contains trials identified from MEDLINE, CINAHL, and the Cochrane Central Register of Controlled Trials (CENTRAL). The search will be conducted using appropriate free text and MeSH terms by adapting terms drawn from the existing search strategies for the Cochrane Colorectal Cancer Group to meet the objectives of this review. MESH headings will include: randomized controlled trial, clinical trial, chemoprevention, adenoma, carcinoma, polyps, colonic, colorectal, nonsteroidal anti‐inflammatory, aspirin, cyclooxygenase inhibitor, cancer. Additional trials will be sought from the UK National Research Register, Controlled Clinical Trials and ZETOC database of conference abstracts (searching specifically the American, European and UK societies of colorectal surgeons’, gastroenterologists’ and specific polyposis meetings abstracts for the past 10 years).

Searching other resources

The reviewers will also search all the reference lists of relevant articles listed in MEDLINE, and will consider handsearching particularly when abstracts and the conference proceeding of associated meetings are not available electronically. We will not impose any language or other limits on the searches. Where there is insufficient information in a published paper an attempt will be made to contact primary authors for clarification or additional information.

Data collection and analysis

Data from RCTs and quasi‐randomised RCTs will be processed as described in the Cochrane Handbook for Systematic Reviews of Interventions (Version 5.1.0, Higgins 2011). Data will be analysed using the RevMan Analyses statistical programme in Review Manager.

Selection of studies

Methodological quality of all included RCT and quasi‐randomised RCT trials in the efficacy review will be evaluated by two reviewers (AL, YM) using the criteria recommended by the Cochrane Handbook to assess risk of bias. In particular, trials will be assessed for randomisation procedure, blinding, adequacy of allocation concealment and intention‐to‐treat analysis. Where this is unclear an attempt will be made to contact primary authors for clarification.

Data extraction and management

Data extraction from the included trials will be undertaken independently by the two review authors (AL, YM). Data from RCTs and quasi‐randomised RCTs will be processed as described in the Cochrane Handbook for Systematic Reviews of Interventions (Version 5.1.0, Higgins 2011). Data will be analysed using the RevMan Analyses statistical programme in Review Manager.

Assessment of risk of bias in included studies

Methodological quality of all included RCT and quasi‐randomised RCT trials will be evaluated by two reviewers using the criteria recommended by the Cochrane Colorectal Review Group to assess risk of bias. In particular, trials will be assessed for randomisation procedure, blinding, adequacy of allocation concealment and intention‐to‐treat analysis.

Measures of treatment effect

We shall report odds ratio (OR), risk ratio (RR) or risk difference (RD) for dichotomous data and weighted mean differences (WMD) for continuous data, accompanied by 95% confidence intervals (CI).

Unit of analysis issues

Studies will be assessed if there are any non‐standard designs.

Dealing with missing data

Trials will be assessed as to whether the results are reported on intention‐to‐treat basis, and if not the number and reasons for withdrawals and drop‐outs will be sought.

Assessment of heterogeneity

On visual inspection of the forest plot together with the chi‐squared test for heterogeneity at the 10% level and the I‐squared statistic if there is evidence of heterogeneity between the trials a reason will be sought by considering the populations, interventions, outcomes and settings of the individual trials. If heterogeneity persists implications will be explored further using random effects models and sensitivity analyses.

Assessment of reporting biases

The potential bias will be reported as published.

Data synthesis

It may be appropriate to use Generic Inverse Variance techniques to combine different scores. If formal synthesis is inappropriate, a narrative overview will be undertaken.

Subgroup analysis and investigation of heterogeneity

If data allow, sub‐group analyses will also be undertaken according to baseline cardiovascular risk and interaction of the study drug, compliance, administration frequency and interaction with other nutritional supplement. Time‐to‐event analysis to draw hazard ratio and a meta regression analysis to elucidate dose‐response relationship will be attempted, if data are available.

Sensitivity analysis

Similarly, if the data allow, sensitivity analysis may be performed to assess the quality of the study data.