You are here

Article Text

Article menu
Colon
Original article
Aspirin use after diagnosis but not prediagnosis improves established colorectal cancer survival: a meta-analysis
  1. Peiwei Li1,2,
  2. Han Wu3,
  3. Honghe Zhang4,
  4. Yu Shi5,
  5. Jinming Xu1,
  6. Yao Ye1,
  7. Dajing Xia1,
  8. Jun Yang5,6,
  9. Jianting Cai2,
  10. Yihua Wu1
  1. 1Department of Epidemiology and Health Statistics, Zhejiang University School of Public Health, Hangzhou, China
  2. 2Department of Gastroenterology, Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
  3. 3The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
  4. 4Department of Pathology, Zhejiang University School of Medicine, Hangzhou, China
  5. 5State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
  6. 6Department of Biomedicine, College of Biotechnology, Zhejiang Agriculture and Forestry University, Hangzhou, China
  1. Correspondence to Dr Yihua Wu, Department of Epidemiology and Health Statistics, Zhejiang University School of Public Health, Hangzhou 310058, China; georgewuer{at}126.com; or Dr Jianting Cai, Department of Gastroenterology, Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310009, China; caijianting111{at}hotmail.com

Abstract

Objective The objective of this meta-analysis was to systematically assess the survival benefit of aspirin use before or after diagnosis for patients with colorectal cancer (CRC).

Design Relevant studies were identified through searching PubMed, Embase and Cochrane databases before May 2014. Two investigators extracted data independently for baseline characteristics and outcomes from the included studies. Either a fixed-effects or a random-effects model was derived to composite the pooled HR for overall mortality and CRC-specific mortality of CRC.

Results Seven studies on postdiagnosis aspirin therapy and seven studies on prediagnosis aspirin use were finally included in this meta-analysis. The overall survival benefit associated with postdiagnosis aspirin use represented an HR of 0.84 (95% CI 0.75 to 0.94). This effect was observed both in colon cancer (HR=0.78, 95% CI 0.64 to 0.96) and in rectal cancer (HR=0.90, 95% CI 0.83 to 0.98). Besides, the survival benefit of postdiagnosis aspirin use appeared to be confined to those patients with positive prostaglandin endoperoxide synthase 2 (PTGS2, also known as cyclooxygenase-2, COX-2) expression (HR=0.65, 95% CI 0.50 to 0.85) and with mutated PIK3CA tumours (HR=0.58, 95% CI 0.37 to 0.90). Aspirin use postdiagnosis was not associated with CRC-specific mortality (HR=0.77, 95% CI 0.52 to 1.14). We observed no evidence of an association between prediagnosis aspirin use and CRC overall mortality (HR=1.01, 95% CI 0.96 to 1.06) or CRC-specific mortality (HR=0.93, 95% CI 0.82 to 1.05).

Conclusions These findings provide further indication that postdiagnosis aspirin therapy improved CRC overall survival, especially for patients with positive PTGS2 (COX-2) expression and mutated PIK3CA tumours.

  • COLORECTAL CANCER

https://doi.org/10.1136/gutjnl-2014-308260

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Significance of this study

    What is already known on this subject?

    • Colorectal cancer (CRC) remains one of the most common types of cancers and a leading cause of death worldwide.

    • Previous observational and randomised data indicated that aspirin reduced the risk of colorectal adenomas and cancer, and risk of distant metastasis for CRC.

    • Several studies reported that aspirin use before or after diagnosis improved CRC survival.

    What are the new findings?

    • Aspirin use after diagnosis decreased all-cause mortality by 16% for patients with CRC.

    • This effect of aspirin use appeared to be confined to those patients with positive prostaglandin endoperoxide synthase 2 (cyclooxygenase-2) expression and patients with mutated PIK3CA tumours.

    • Prediagnosis aspirin use is not associated with CRC overall mortality or CRC-specific mortality.

    How might it impact on clinical practice in the foreseeable future?

    • For patients with CRC, aspirin use after diagnosis may be considered to be a low-cost therapy to improve overall survival.

    Introduction

    Colorectal cancer (CRC) remains one of the most common type of cancers and a leading cause of death worldwide.1 CRC accounts for about 8% of annual cancer-related deaths overall, with over 1.2 million new cancer cases and 608 700 deaths estimated to have occurred in 2008.2 Although morbidity and mortality from CRC have decreased, and systemic therapy of CRC over the last decades has improved modestly, there is still an urgent need for more effective preventive strategies and adjuvant therapies against CRC.3

    A significant body of observational and randomised data has indicated that aspirin has anticancer effects.4–9 Studies over the past decades have suggested that aspirin use could reduce the risk of colorectal adenomas and cancer.10 In addition, a meta-analysis of randomised trials designed to evaluate the cardiovascular benefits of aspirin demonstrated that aspirin use reduced the risk of distant metastasis for patients with CRC.11 Nevertheless, it remains uncertain whether aspirin use can influence the prognosis for established patients with CRC. Recently, several studies assessed the effect of postdiagnosis or prediagnosis aspirin use on survival of established patients with CRC.3 ,6–9 ,12–17 Although it remains unclear for the optimal time, dose and duration of aspirin use and which type of patients with CRC are most likely to benefit, these data suggest aspirin to be a promising adjuvant therapy for patients with CRC. In this meta-analysis, we assessed the survival benefit of prediagnosis or postdiagnosis aspirin use for patients with CRC. Moreover, we investigated whether the effect of aspirin use differed according to factors such as localisation of tumours and tumour biomarkers. This may provide further insight into the anticancer mechanisms of aspirin and indicate evidence for aspirin therapy in patients with CRC.

    Methods

    Literature search and study selection

    The protocol for this systematic review was based on the PRISMA statement.18 We performed systematic literature searches of PubMed, Embase and Cochrane databases before May 2014 for possible publications. Reports cited the references identified in this systematic review and relevant reviews were also searched to include potentially missed studies. The following terms were used in the search procedure: (‘colorectal cancer’ or ‘colon cancer’ or ‘rectal cancer’ or ‘colorectal adenocarcinoma’ or ‘colon adenocarcinoma’ or ‘rectal adenocarcinoma’) AND (‘aspirin’ or ‘non-steroidal anti-inflammatory drugs’ or ‘NSAIDS’). The retrieved studies were carefully examined to exclude potential duplicates or overlapping data. Titles and abstracts of articles selected from the initial search were first scanned, and then full papers of potential eligible studies were reviewed.

    Eligibility of studies for inclusion was assessed independently by two investigators. Studies were eligible for inclusion if all the following criteria were fulfilled: the outcome of interest was survival of colorectal, colon or rectal cancer; the study of interest was aspirin usage prediagnosis and/or postdiagnosis of CRC; the HR or risk ratio (RR) estimates with 95% CIs were available. If the data sets overlapped or were duplicated, we only extracted the most detailed or recent information. Only studies published in English were included.

    Data extraction

    Data were extracted independently by two authors, and any discrepancies were resolved by a third investigator. The endpoints of this analysis were all-cause mortality and CRC-specific mortality, in accordance with most of the included studies. Participant characteristics and outcomes were extracted from each study, and the principal author or the corresponding author was contacted if the required data were unavailable. HRs (RRs) reflecting the greatest degree of control for potential confounders were adopted in this meta-analysis. An improved Newcastle-Ottawa Scale was applied to assess the study quality.19 Studies were categorised as high quality if the score was 9 points or more, as medium quality if the score was 5–8 points and as low quality if the score was <5.

    Statistical analysis

    Heterogeneity between individual studies was assessed by χ2 test and I2 test; p≤0.05 and/or I2>50% indicates significant heterogeneity.20 Summary HRs (RRs) and 95% CI were calculated using a random-effects model when the heterogeneity was significant, and a fixed-effects model was applied otherwise. The primary meta-analyses were conducted to assess the efficacy of prediagnosis or postdiagnosis aspirin usage on overall mortality and CRC-specific mortality of patients with CRC. Subgroup analyses were further conducted for the association between postdiagnosis aspirin use and overall mortality, stratifying by anatomical sites and molecular biomarker status of CRC. Sensitivity and subgroup analyses were used to dissect the heterogeneity. To evaluate the publication bias risk, funnel plots were evaluated. We did not conduct further statistical tests for funnel plot asymmetry because of the limited test power when fewer than 10 studies are included.21 Two-sided p values were calculated, with a p value <0.05 considered significant for all tests. All analyses were conducted using the Stata software (V.11.0; StataCorp, College Station, Texas, USA).

    Results

    Description of the included studies

    Our systematic literature search identified 1617 articles for eligibility, of which 42 were potentially relevant reports for further review. After excluding 31 papers, a total of 11 articles met the inclusion criteria for the final meta-analyses (figure 1).3 ,6–9 ,12–17 Two articles reported the same studies (Nurses’ Health Study and Health Professionals Follow-up Study),9 ,14 and thus data from Liao et al14 were used only in the subgroup analysis of PIK3CA mutation and CRC overall mortality. A total of seven studies comprising 34 933 patients with CRC reported the survival benefit of postdiagnosis aspirin use for CRC (all assessed overall mortality and three investigated CRC-specific mortality). And seven studies with 30 360 patients evaluated the association between prediagnosis aspirin use and CRC survival (five on overall mortality and six reported CRC-specific mortality).

    Figure 1
    Figure 1

    Flow diagram of the selection process. Several studies assessed both prediagnosis and postdiagnosis aspirin use and outcomes (all-cause mortality and colorectal cancer-specific mortality).

    The characteristics of the included studies are shown in table 1. All the studies were conducted in the USA and in European countries. Nine of the included studies were cohort studies, including three population-based cohorts9 ,14 ,17 and five cancer registry studies6 ,7 ,13 ,15 ,16; one study was nested cohort within randomised controlled trial (RCT),3 one was nested case–control study within cohort12 and the other one was a case–control study.8 Sample size of the included studies ranged from 621 to 13 944. One study enrolled patients with colon cancer only,6 while the other studies included both the patients with colon and rectal cancer. Most of the included studies investigated patients with CRC of all stages (I–IV), while one study enrolled only stage I–III patients9 and another reported II–III stage patients.3 The methodological quality of the 11 included studies was generally moderate to good, as shown in online supplementary table S1.

    View this table:
    Table 1

    Characteristics of the studies evaluating the survival benefit of aspirin use for patients with CRC

    Aspirin use after diagnosis and CRC survival

    Seven studies assessed the overall survival benefit associated with aspirin use after diagnosis of CRC. In total, 8666 patients died from any cause in 25 563 aspirin non-users and 2977 patients died in 9370 aspirin users. The pooled analysis represented an HR of 0.84 (95% CI 0.75 to 0.94) with significant heterogeneity (I2=78.2%, p<0.001) (figure 2A). In the three studies evaluating CRC-specific mortality, there were 1906 CRC-specific deaths among 9091 participants who did not use aspirin after CRC diagnosis compared with 703 CRC-specific deaths in 3821 aspirin users. The pooled HR was 0.77 (95% CI 0.52 to 1.14) for CRC-specific mortality with significant heterogeneity across studies (I2=89.1%, p<0.001) (figure 2B). After excluding the study applying case–control analysis of a cohort study,12 the heterogeneity decreased for all-cause mortality (I2=67.9%, p=0.008) and CRC-specific mortality (I2=4.3%, p=0.307).

    Figure 2
    Figure 2

    Meta-analysis of the association between aspirin use and colorectal cancer (CRC) mortality. (A) Postdiagnosis aspirin use and all-cause mortality. (B) Postdiagnosis aspirin use and CRC-specific mortality. (C) Prediagnosis aspirin use and all-cause mortality. (D) Prediagnosis aspirin use and CRC-specific mortality.

    The results of subgroup analyses for the association between postdiagnosis aspirin use and CRC overall mortality are demonstrated in table 2. When the analysis was stratified according to colon and rectal subsites, the pooled HRs of colon cancer and rectal cancer were 0.78 (95% CI 0.64 to 0.96) and 0.90 (95% CI 0.83 to 0.98), respectively. The results showed that aspirin use after diagnosis reduced all-cause death for both patients with colon and rectal cancer. The effect of PIK3CA mutation and prostaglandin endoperoxide synthase 2 (PTGS2, also known as cyclooxygenase-2, COX-2) expression on the survival benefit associated with postdiagnosis aspirin use was also examined. The benefit of postdiagnosis aspirin use appeared to be confined to participants with mutated-PIK3CA tumours (pooled HR=0.58, 95% CI 0.37 to 0.90) but not for wild-PIK3CA patients (pooled HR=0.78, 95% CI 0.53 to 1.14) (figure 3A, C). For PTGS2 (COX-2) expression status, use of aspirin after diagnosis was significantly associated with a lower overall mortality when tumours presented strong PTGS2 (COX-2) expression (pooled HR=0.65, 95% CI 0.50 to 0.85) but not for tumours with weak PTGS2 (COX-2) expression (pooled HR=0.75, 95% CI 0.43 to 1.30) (figure 3B, D).

    View this table:
    Table 2

    Subgroup analysis for the association between postdiagnosis aspirin usage and all-cause mortality of CRC

    Figure 3
    Figure 3

    Meta-analysis of the association between postdiagnosis aspirin use and all-cause mortality, according to PTGS2 (COX-2) expression and PIK3CA mutation status. (A) Overall mortality among patients with muted- PIK3CA tumours. (B) Overall mortality among patients with strong PTGS2 (COX-2) expression tumours. (C) Overall mortality among patients with wild-type PIK3CA tumours. (D) Overall mortality among patients with weak PTGS2 (COX-2) expression tumours. COX-2, cyclooxygenase-2; PTGS2, prostaglandin endoperoxide synthase 2.

    Aspirin use prediagnosis and CRC survival

    Five studies comprising 21 093 patients with CRC and 8734 all-cause deaths evaluated the effect of prediagnosis aspirin use on CRC overall mortality, while six studies with 16 416 patients with CRC and 3902 CRC-specific deaths assessed CRC-specific mortality. There was no evidence of an association between prediagnosis aspirin use and CRC overall mortality (pooled HR=1.01, 95% CI 0.96 to 1.06) or CRC-specific mortality (pooled HR=0.93, 95% CI 0.82 to 1.05) (figure 2C, D).

    Publication bias

    No evidence of publication bias was indicated in this study. However, because of the limited number of included studies, whether the publication bias exists in the current meta-analysis is still difficult to confirm.

    Discussion

    The current meta-analysis summarised the results of eleven studies, including seven studies on aspirin use after CRC diagnosis and seven studies on aspirin use before CRC diagnosis. The results indicated that use of aspirin postdiagnosis was inversely associated with the overall mortality of CRC. This association appeared to be slightly stronger among patients with colon cancer than among patients with rectal cancer (HR=0.78 for colon cancer and 0.90 for rectal cancer). Besides, the present meta-analysis further suggested that increased expression of PTGS2 (COX-2) and the presence of mutated PIK3CA in the primary tumours served as predictive molecular biomarkers for adjuvant aspirin therapy.

    There has been high-quality evidence showing a benefit of aspirin use in sporadic adenomatous polyps prevention, and observational studies have suggested aspirin to be associated with a reduction in CRC risk and mortality.22 In addition to the well-recognised chemoprevention effect against CRC, recent data also suggest a role of aspirin as an adjuvant agent for patients with CRC. However, the precise molecular mechanisms underlying the effects of aspirin against CRC are still unclear. Reimers et al6 speculated that the effects of aspirin on CRC survival might result from an effect on circulating tumour cells and their ability to develop into metastatic cancer. This is also supported by the pervious evidence from RCTs, which indicated that aspirin use reduced the risk of cancer metastasis.11 As suggested by a previous study, direct platelet–tumour cell contacts may activate the tumour growth factor-β/SMAD and nuclear factor-κB (NFκB) pathways in cancer cells, resulting in an invasive mesenchymal-like phenotype with metastatic potential.23 Moreover, a molecular study indicated that acquisition of an epithelial–mesenchymal phenotype markedly reduced inhibition of cytotoxic T-lymphocytes-mediated tumour cell lysis.24 Thus, aspirin may prevent cancer metastasis through inhibiting platelet–tumour cell signalling and epithelial–mesenchymal transition in circulating tumour cells.

    The anticancer effect of aspirin has also been attributed to direct inhibition of PTGS (COX) family of enzymes involved in prostaglandin synthesis.25 Previous studies have investigated the interaction between aspirin and tumour PTGS2 expression and tumour PIK3CA mutation.3 ,6 ,9 ,14 These molecular pathological epidemiology (MPE) studies were well-designed cohort studies and provided further evidence for the role of PIK3CA and PTGS2 (COX-2) in CRC progression, while they were all conducted in European countries and in the USA, and the PIK3CA mutation and PTGS2 overexpression rate differed across the studies. Thus, further studies, especially prospective cohorts or RCTs with large sample size, are still warranted to validate the findings and to investigate this issue in other ethnic groups such as Asians. Though there are some limitations in MPE studies such as selection bias, confounding and causal inference, this evolving field of epidemiology could provide new insights into the pathogenic processes and optimise personalised prevention or treatment of CRC.26 PTGS2 (COX-2) was indicated to promote inflammation and cell proliferation, and overexpression of PTGS2 (COX-2) has been observed to be associated with poor prognosis of CRC.27 ,28 The phosphatidylinositol 3-kinase (PI3K) signalling pathway plays a vital role in carcinogenesis, and upregulation of PI3K enhances PTGS2 (COX-2) activity and prostaglandin E2 (PGE2) synthesis.29 ,30 These findings might partially explain why the effect of aspirin on survival among patients with muted PIK3CA and PTGS2 (COX-2)-positive tumours differs from those patients with wild-type PIK3CA and PTGS2 (COX-2)-negative tumours. Aspirin also has a broader range of downstream effect, including inhibition of NFκB,31 induction of apoptosis by activation of p38 kinase32 and catabolism of polyamines.33

    There are still some confusing problems unresolved, including when aspirin should be used and what is the optimal dose and duration for aspirin adjuvant therapy. Our analysis further supported aspirin use after diagnosis but not prior diagnosis improved survival of CRC. While significant data suggested that aspirin has a benefit as a chemoprevention agent for CRC, we should consider whether its adverse effect such as GI tract bleeding outweighs its benefit. Thus, it is critical to identify the lowest dose of aspirin that can achieve sufficient antitumour effect because of the potential toxicities of aspirin. The included observational studies adopted dosages between 75 and more than 300 mg daily. Pharmacological data indicated that low-dose aspirin use at 75–100 mg daily was sufficient to completely inhibit PTGS1 in platelets, and 81 mg aspirin is sufficient to inhibit rectal mucosal PGE2 production.34 ,35 Several studies have suggested that regular aspirin use may be associated with improved survival.13 ,15 An ongoing randomised study is conducted in high-risk Dukes B and C patients with CRC to evaluate the survival benefit for a dose of 200 mg aspirin daily (ASCOLT).36 Because of limited data, dose–response analysis for the association between aspirin use and CRC survival could not be conducted and the optimal duration of aspirin therapy is still unclear.

    Our meta-analysis comprehensively analysed observational data on the survival benefit of aspirin therapy for established patients with CRC, and the results further suggested that aspirin use after diagnosis improved CRC survival. This may be important for clinical practice of CRC; however, the current analysis is restricted by several limitations. First, the number of studies involved in this meta-analysis was relatively small, and thus some of the subgroup analyses were difficult to conduct and might be less reliable. Second, the dose and duration of aspirin usage differed among studies and related data of individuals were not available from each study, which may possibly result in less accurate estimates of survival benefit for aspirin therapy.

    Third, the heterogeneity across studies could not be fully explained. The sensitivity analysis indicated that the meta-analyses were influenced by individual study. After excluding the study with case–control design,12 heterogeneity decreased for meta-analyses of both all-cause mortality and CRC-specific mortality. Other factors, including aspirin dose, duration, sample size and distribution of tumour stages, may also be the source of heterogeneity. Therefore, further analysis should be conducted if individual participant data are available. Fourth, the available studies involved in the current analysis were mostly observational studies, which are vulnerable to selection bias, information bias and confounding. Thus, inadequate control of the confounders might lead to exaggeration or underestimation of survival benefit estimates. Besides, all the studies were performed in the USA or in European countries.

    In conclusion, the results from this meta-analysis demonstrated that aspirin use after diagnosis was inversely associated with CRC overall mortality, particular among individuals with positive PTGS2 (COX-2) expression and mutated PIK3CA tumours. This highlights that PTGS2 (COX-2) and PIK3CA are potential markers to tailor aspirin use for patients with CRC. The findings have considerable clinical and health policy importance since aspirin is cheap and may provide a cost-effective approach to improve CRC survival. If finally proved to be effective, aspirin might challenge the paradigm of adjuvant therapy for CRC. Future research, especially RCT, is desirable to confirm this low-cost and easily administered, relatively safe therapy to reduce CRC mortality and to enable a better understanding of the role of aspirin adjuvant therapy for CRC survival.

    Reference

    1. ↵
      1. Siegel R,
      2. Ma J,
      3. Zou Z, et al
      . Cancer statistics, 2014. CA Cancer J Clin 2014;64:9–29.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Jemal A,
      2. Bray F,
      3. Center MM, et al
      . Global cancer statistics. CA Cancer J Clin 2011;61:69–90.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Domingo E,
      2. Church DN,
      3. Sieber O, et al
      . Evaluation of PIK3CA mutation as a predictor of benefit from nonsteroidal anti-inflammatory drug therapy in colorectal cancer. J Clin Oncol 2013;31:4297–305.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Algra AM,
      2. Rothwell PM
      . Effects of regular aspirin on long-term cancer incidence and metastasis: a systematic comparison of evidence from observational studies versus randomised trials. Lancet Oncol 2012;13:518–27.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Langley RE,
      2. Burdett S,
      3. Tierney JF, et al
      . Aspirin and cancer: has aspirin been overlooked as an adjuvant therapy? Br J Cancer 2011;105:1107–13.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Reimers MS,
      2. Bastiaannet E,
      3. Langley RE, et al
      . Expression of HLA class I antigen, aspirin use, and survival after a diagnosis of colon cancer. JAMA Intern Med 2014;174:732–9.
      OpenUrlCrossRefPubMed
    7. ↵
      1. McCowan C,
      2. Munro AJ,
      3. Donnan PT, et al
      . Use of aspirin post-diagnosis in a cohort of patients with colorectal cancer and its association with all-cause and colorectal cancer specific mortality. Eur J Cancer 2013;49:1049–57.
      OpenUrlCrossRefPubMed
    8. ↵
      1. Din FVN,
      2. Theodoratou E,
      3. Farrington SM, et al
      . Effect of aspirin and NSAIDs on risk and survival from colorectal cancer. Gut 2010;59:1670–9.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Chan AT,
      2. Ogino S,
      3. Fuchs CS
      . Aspirin use and survival after diagnosis of colorectal cancer. JAMA 2009;302:649–58.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Dube C,
      2. Rostom A,
      3. Lewin G, et al
      . The use of aspirin for primary prevention of colorectal cancer: a systematic review prepared for the U.S. Preventive Services Task Force. Ann Intern Med 2007;146:365–75.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Rothwell PM,
      2. Wilson M,
      3. Price JF, et al
      . Effect of daily aspirin on risk of cancer metastasis: a study of incident cancers during randomised controlled trials. Lancet 2012;379:1591–601.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Cardwell CR,
      2. Kunzmann AT,
      3. Cantwell MM, et al
      . Low-dose aspirin use after diagnosis of colorectal cancer does not increase survival: A case-control analysis of a population-based cohort. Gastroenterology 2014;146:700–8.e2.
      OpenUrlCrossRefPubMed
    13. ↵
      1. Walker AJ,
      2. Grainge MJ,
      3. Card TR
      . Aspirin and other non-steroidal anti-inflammatory drug use and colorectal cancer survival: a cohort study. Br J Cancer 2012;107:1602–7.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Liao X,
      2. Lochhead P,
      3. Nishihara R, et al
      . Aspirin use, tumor PIK3CA mutation, and colorectal-cancer survival. New Engl J Med 2012;367:1596–606.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Bastiaannet E,
      2. Sampieri K,
      3. Dekkers OM, et al
      . Use of Aspirin postdiagnosis improves survival for colon cancer patients. Br J Cancer 2012;106:1564–70.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Coghill AE,
      2. Newcomb PA,
      3. Campbell PT, et al
      . Prediagnostic non-steroidal anti-inflammatory drug use and survival after diagnosis of colorectal cancer. Gut 2011;60:491–8.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Zell JA,
      2. Ziogas A,
      3. Bernstein L, et al
      . Nonsteroidal anti-inflammatory drugs: effects on mortality after colorectal cancer diagnosis. Cancer 2009;115:5662–71.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Moher D,
      2. Liberati A,
      3. Tetzlaff J, et al
      . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097.
      OpenUrlCrossRefPubMed
    19. ↵
      1. Stang A
      . Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010;25:603–5.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Higgins JP,
      2. Thompson SG
      . Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539–58.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Sterne JA,
      2. Sutton AJ,
      3. Ioannidis JP, et al
      . Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ 2011;343:d4002.
      OpenUrlFREE Full Text
    22. ↵
      1. Thun MJ,
      2. Jacobs EJ,
      3. Patrono C
      . The role of aspirin in cancer prevention. Nat Rev Clin Oncol 2012;9:259–67.
      OpenUrlCrossRefPubMed
    23. ↵
      1. Labelle M,
      2. Begum S,
      3. Hynes RO
      . Direct signaling between platelets and cancer cells induces an epithelial-mesenchymal-like transition and promotes metastasis. Cancer Cell 2011;20:576–90.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Akalay I,
      2. Janji B,
      3. Hasmim M, et al
      . Epithelial-to-mesenchymal transition and autophagy induction in breast carcinoma promote escape from T-cell-mediated lysis. Cancer Res 2013;73:2418–27.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Chan AT,
      2. Ogino S,
      3. Fuchs CS
      . Aspirin and the risk of colorectal cancer in relation to the expression of COX-2. N Engl J Med 2007;356:2131–42.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Ogino S,
      2. Chan AT,
      3. Fuchs CS, et al
      . Molecular pathological epidemiology of colorectal neoplasia: an emerging transdisciplinary and interdisciplinary field. Gut 2011;60:397–411.
      OpenUrlAbstract/FREE Full Text
    27. ↵
      1. Dong M,
      2. Johnson M,
      3. Rezaie A, et al
      . Cytoplasmic phospholipase A2 levels correlate with apoptosis in human colon tumorigenesis. Clin Cancer Res 2005;11: 2265–71.
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Soumaoro LT,
      2. Uetake H,
      3. Higuchi T, et al
      . Cyclooxygenase-2 expression: a significant prognostic indicator for patients with colorectal cancer. Clin Cancer Res 2004;10:8465–71.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Samuels Y,
      2. Wang Z,
      3. Bardelli A, et al
      . High frequency of mutations of the PIK3CA gene in human cancers. Science 2004;304:554.
      OpenUrlFREE Full Text
    30. ↵
      1. Kaur J,
      2. Sanyal SN
      . PI3-kinase/Wnt association mediates COX-2/PGE(2) pathway to inhibit apoptosis in early stages of colon carcinogenesis: chemoprevention by diclofenac. Tumour Biol 2010;31:623–31.
      OpenUrlCrossRefPubMed
    31. ↵
      1. Kopp E,
      2. Ghosh S
      . Inhibition of NF-kappa B by sodium salicylate and aspirin. Science 1994;265:956–9.
      OpenUrlAbstract/FREE Full Text
    32. ↵
      1. Schwenger P,
      2. Bellosta P,
      3. Vietor I, et al
      . Sodium salicylate induces apoptosis via p38 mitogen-activated protein kinase but inhibits tumor necrosis factor-induced c-Jun N-terminal kinase/stress-activated protein kinase activation. Proc Natl Acad Sci USA 1997;94:2869–73.
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Martinez ME,
      2. O'Brien TG,
      3. Fultz KE, et al
      . Pronounced reduction in adenoma recurrence associated with aspirin use and a polymorphism in the ornithine decarboxylase gene. Proc Natl Acad Sci USA 2003;100:7859–64.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Ruffin MT IV.,
      2. Krishnan K,
      3. Rock CL, et al
      . Suppression of human colorectal mucosal prostaglandins: determining the lowest effective aspirin dose. J Natl Cancer Inst 1997;89:1152–60.
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Sample D,
      2. Wargovich M,
      3. Fischer SM, et al
      . A dose-finding study of aspirin for chemoprevention utilizing rectal mucosal prostaglandin E(2) levels as a biomarker. Cancer Epidemiol Biomarkers Prev 2002;11:275–9.
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Ali R,
      2. Toh HC,
      3. Chia WK
      . The utility of Aspirin in Dukes C and High Risk Dukes B Colorectal cancer—the ASCOLT study: study protocol for a randomized controlled trial. Trials 2011;12:261.
      OpenUrlCrossRefPubMed
    View Abstract

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • PL and HW contributed equally.

    • Contributors YW and JC contributed to conception and design of the study. PL, HW and HZ contributed to conception, design of the study and editing the manuscript. YS, JX and YY contributed to statistical analysis. DX and JY contributed to the data acquisition, analysis and interpretation of the data. All authors commented on drafts of the paper and have approved the final draft of the manuscript.

    • Funding The work was funded by the National Natural Science Foundation of China (grant nos. 81302455, 81201557 and 81373036).

    • Competing interests None.

    • Provenance and peer review Not commissioned; externally peer reviewed.