Background

Esophageal, gastric and colorectal cancers are the most common cancers of the digestive tract [1]. Many factors, including old age and poor living habits, are risk factors for gastrointestinal malignancies. Although the incidence and mortality of gastrointestinal malignancies have been reduced in recent years, the comprehensive treatment of gastrointestinal malignancies has progressed slowly in recent decades. Therefore, it is urgent to discover a more effective comprehensive treatment for gastrointestinal malignancies. Aspirin is a nonselective cyclooxygenase inhibitor with strong antipyretic and analgesic effects and is widely used for its anti-inflammatory and anti-rheumatic properties. For example, small doses of aspirin are used to prevent the onset of cardiovascular disease, cerebrovascular disease and transient ischemic attacks. In recent years, many studies [2,3,4,5,6,7] have found that aspirin also has anticancer effects. However, as there are still some controversy about these studies, the aim of this study was to assess the survival benefits of aspirin use (compared with non-aspirin use) for esophageal, gastric and colorectal cancer patients through the use of a meta-analysis.

Methods

All of the search results were evaluated according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2009) statement [8].

Inclusion and exclusion criteria

The inclusion criteria were as follows: (1) RCTs or observational studies including cohort and case-control studies; (2) the outcomes of interest beingdefined as OS (overall survival) and CSS (cancer-specific survival) of esophageal, gastric, colorectal, colon or rectal cancer; (3) the study addressing aspirin usage at the times of prediagnosis and/or postdiagnosis of esophageal, gastric, colorectal, colon or rectal cancer; (4) HR or OR estimates with 95% CIs were available. The exclusion criteria were as follows: (1) duplicate articles; (2) inadequate data; and (3) sample sizes less than 20; (4) NOS ≤5.

Literature search

We conducted a comprehensive systematic literature search of online databases, including PubMed, the Cochrane Library, Embase and www.clinicaltrials.gov for studies that were conducted before April 30th, 2020, to identify all RCTs and observational studies. The following key words were used in these literature searches: (‘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’) (‘gastric cancer’ or ‘gastric adenocarcinoma’) AND (‘aspirin’ or ‘non-steroidal anti-inflammatory drugs’ or ‘NSAIDS’) (‘esophageal cancer’ or ‘esophageal adenocarcinoma’ or ‘esophageal squamous cell carcinoma’) AND (‘aspirin’ or ‘non-steroidal anti-inflammatory drugs’ or ‘NSAIDS’). There were no language restrictions. We also reviewed the references of the included articles and of the related systematic reviews, in order to identify additional studies.

Study selection and quality assessment

The qualities of the included non-RCTs were assessed by using the Newcastle–Ottawa Scale (NOS) [9]. The scale utilizes a score system ranging from 0 to 9, and the quality of the observational studies were considered to be high-quality with a score of 5 or higher.

Data extraction

Data extraction and the evaluation of the quality of the literature were independently conducted by 2 investigators (Ju-li Lin and Jian-xian Lin). At time when there was any uncertainty about the inclusion of a study, the issue was discussed between the two investigators to achieve a resolution. A Microsoft Excel database was employed to record all of the available information, including the baseline details, title, first author’s name, year of publication, study design, region, journal, sample size, period of patient recruitment, follow-up time, and HRs.

Statistical analysis

The Cochran’s Q statistic and I2 statistics were applied to assess the heterogeneity among all of the studies [10]. For the Q statistic, a p value of less than 0.1 was considered to be statistically significant. When statistical heterogeneity was detected, the sources of the heterogeneity were explored, and sensitivity analyses were performed. A random-effects model was used if heterogeneity existed; otherwise, the fixed-effect model was used. When possible, subgroup analyses were conducted to assess the potential impacts of the mutation statuses. The cut-off point for quality among observational studies (NOS ≤5 vs. NOS > 5) was arbitrarily defined. Publication bias was assessed using the Begg and Egger regression asymmetry test, together with funnel plots. All of the statistical analyses were conducted by using STATA, version 13.0 (Stata Corporation, College Station, TX).

Results

Retrieved studies and characteristics

According to the previously described search strategy, 3612 citations were obtained from the online database up until April 30th, 2020. A total of 3569 articles were excluded by viewing the titles and abstracts. The full texts of 36 records were read. Ultimately, 18 full-text studies [4,5,6,7, 11,12,13,14,15,16,17,18,19,20,21,22,23,24] were obtained and assessed according to the eligibility criteria, including 1 case-control study and 17 cohort studies, with the studies comprised of more than 74,936 patients. The detailed literature search and screening process are shown in Supplement Figure 1. The characteristics included in the study are shown in Tables 1 and 2, including the first author’s name, year of publication, study design, region, journal, sample size, period of patient recruitment patients, follow-up time and definition of aspirin use.

Table 1 Characteristics of the included trials and particiants
Full size table
Table 2 Characteristics of the included trials and particiants
Full size table

The qualities of 18 studies was assessed by using NOS; four studies achieved a score of 6, six studies achieved a score of 7 and eight studies achieved a score of 8 (Tables 3 and 4). Thirteen studies stated a clear follow-up time. The longest median follow-up period was 10.8 years. Six studies reported a clear definition of the use of PPIs. Seven studies compared the risk of gastric cancer between PPI users and non-PPI users. Thirteen studies evaluated the association between prediagnosis aspirin use and colorectal cancer survival. Thirteen studies evaluated the association between postdiagnosis aspirin use and colorectal cancer survival.

Table 3 Quality assessment of the observational studies using the Newcastle-Ottawa Scale (NOS). Assessment of the cohort studies
Full size table
Table 4 Quality assessment of the observational studies using the Newcastle-Ottawa Scale (NOS). Assessment of the case–control study
Full size table

Association between postdiagnosis aspirin use and survival (OS and CSS) in esophageal and gastric cancers

Three studies (involving 6797 patients) compared the overall survival of esophageal cancer among aspirin users compared with non-aspirin users. The estimated pooled HRs showed no significant differences between the two groups [HR = 1.009, 95%CI(0.847, 1.202)] (Fig. 1a).

Fig. 1
figure 1

a post-diagnosis aspirin use and overall survival for upper digestive cancer. b post-diagnosis aspirin use and cancer specific survival for upper digestive cancer

Full size image

Two studies (involving 4589 patients) compared the overall survival of gastric cancer among aspirin users compared with non-aspirin users, and the estimated pooled HRs indicated no significant differences between the groups [HR = 0.870, 95%CI(0.470, 1.610)] (Fig. 1a).

Three studies (involving 11,380 patients) compared the overall survival of upper digestive cancer among aspirin users compared with non-aspirin users, with no significant differences between the two groups based on estimated pooled HRs [HR = 0.831, 95%CI(0.679, 1.016)] (Fig. 1a).

One study (involving 946 patients) compared the cancer-specific survival of esophageal cancer among aspirin users with non-aspirin users; based on HRs, the use of aspirin postdiagnosis was associated with longer cancer-specific survival [HR = 0.34, 95%CI(014, 0.69)] (Fig. 1b). One study involving 750 patients compared the cancer-specific survival of gastric cancer among aspirin users with non-aspirin users, and the HRs revealed no significant differences between the groups [HR = 0.70, 95% CI (0.29, 1.69)] (Fig. 1b).

Association between postdiagnosis aspirin use and survival (OS and CSS) in colorectal cancer

Ten studies (involving 67,552 patients) compared the overall survival of colorectal cancer among aspirin users compared with non-aspirin users. According to the estimated pooled HRs, the use of aspirin postdiagnosis was associated with longer overall survival [HR = 0.83, 95%CI(0.75, 0.93)] (Fig. 2a).

Fig. 2
figure 2

a post-diagnosis aspirin use and overall survival for colorectal cancer.b cumulative meta-analysis of the HR for the colorectal cancer according to time. c post-diagnosis aspirin use and cancer specific survival for colorectal cancer. d cumulative meta-analysis of the HR for the colorectal cancer according to time

Full size image

The result of cumulative meta-analysis showed that the significant difference supporting PPI use was first found in the latest study in Joseph et al. [HR = 0.89, 95% CI(0.86–0.93)], with the CI narrowing and the effect size becoming stable (Fig. 2b).

Eight studies (involving 52,662 patients) compared cancer-specific survival in colorectal cancer among aspirin users and non-aspirin users. The estimated pooled HRs showed that the use of aspirin postdiagnosis was associated with longer overall survival [HR = 0.78, 95%CI(0.66, 0.92)] (Fig. 2c).

The result of cumulative meta-analysis indicated that the significant difference supporting PPI use was first found in the latest study by Joseph et al. [HR = 0.85, 95% CI (0.80–0.89], with the CI narrowing and the effect size becoming stable (Fig. 2d).

Association between prediagnosis aspirin use and survival (OS and CSS) in colorectal cancer

With regard to overall survival in colorectal cancer, five studies involving 6202 patients compared among aspirin users compared with non-aspirin users. The estimated pooled HRs demonstrated no significant differences between the two groups [HR = 1.01, 95%CI(0.96, 1.06)] (Fig. 3a).

Fig. 3
figure 3

a pre-diagnosis aspirin use and overall survival for colorectal cancer. b pre-diagnosis aspirin use and cancer specific survival for colorectal cancer

Full size image

Five studies (involving 45,101 patients) compared the cancer-specific survival of colorectal cancer among aspirin users compared with non-aspirin users, and according to the estimated pooled HRs, there were no significant differences between the groups [HR = 0.93, 95%CI(0.84, 1.03)] (Fig. 3b).

Association between both prediagnosis and postdiagnosis aspirin use and survival (OS and CSS) in colorectal cancer

Four studies (involving 2350 patients) compared the overall survival of colorectal cancer among aspirin users compared with non-aspirin users. The estimated pooled HRs revealed that the use of aspirin both prediagnosis and postdiagnosis was associated with longer overall survival [HR = 0.75, 95%CI(0.61, 0.92)] (Fig. 4a).

Fig. 4
figure 4

a both pre and post-diagnosis aspirin use and overall survival for colorectal cancer. b both pre and post -diagnosis aspirin use and cancer specific survival for colorectal cancer

Full size image

Three studies (involving 1849 patients) compared cancer-specific survival in colorectal cancer among aspirin users compared with non-aspirin users, and the estimated pooled HRs indicated that the use of aspirin both prediagnosis and postdiagnosis was associated with longer overall survival [HR = 0.78, 95%CI(0.73, 0.85)] (Fig. 4b).

Subgroup analysis according to the PIK3CA gene status

Four studies (involving 4346 patients) compared the overall survival of colorectal cancer among aspirin users compared with non-aspirin users among those with PIK3CA gene mutation. Based on the estimated pooled HRs, the use of aspirin postdiagnosis was associated with longer overall survival [HR = 0.70, 95%CI(0.50, 0.99)] (Fig. 5a).

Fig. 5
figure 5

a post-diagnosis aspirin use and overall survival for colorectal cancer according to PIK3CA mutation. b post-diagnosis aspirin use and cancer specific survival for colorectal cancer according to PIK3CA mutation. c post-diagnosis aspirin use and overall survival for colorectal cancer according to PTGS2(COX-2) mutation

Full size image

For overall survival in colorectal cancer, three studies involving 8490 patients compared among aspirin users compared with non-aspirin users among patients with a wild-type PIK3CA gene, and the estimated pooled HRs showed no significant differences between the groups [HR = 0.79, 95%CI(0.53, 1.13)] (Fig. 5a).

Two studies involving 2451 patients compared the cancer-specific survival in colorectal cancer among aspirin users compared with non-aspirin users among patients with a mutated PIK3CA gene. The estimated pooled HRs showed that the use of aspirin postdiagnosis was associated with longer overall survival [HR = 0.27, 95%CI(0.08, 0.91)] (Fig. 5b).

Subgroup analysis according to the PTGS2 (COX-2) expression status

Two studies involving 560 patients compared overall survival in colorectal cancer among aspirin users compared with non-aspirin users in patients with strong PTGS2 (COX-2) expression. According to the estimated pooled HRs, the use of aspirin postdiagnosis was associated with longer overall survival [HR = 0.65, 95%CI(0.54, 0.83)] (Fig. 5c).

Regarding the overall survival of colorectal cancer, two studies involving 4328 patients compared aspirin users with non-aspirin users among patients with weak PTGS2 (COX-2) expression. The estimated pooled HRs showed no significant differences between the two groups [HR = 0.75, 95%CI(0.43, 1.30)] (Fig. 5c).

Subgroup analysis according tumor stage

Four studies involving 28,032 patients compared overall survival in colorectal cancer among aspirin users compared with non-aspirin users among patients. The estimated pooled HRs showed no significant differences between the groups (Supplement Figure 3A).

Five studies involving 32,826 patients compared cancer specific survival in colorectal cancer among aspirin users compared with non-aspirin users. The estimated pooled HRs showed no significant differences between the groups in stage I, stage III and stage IV patients. While the use of aspirin was associated with longer cancer specific survival in stage II patients [HR = 0.65, 95%CI(0.54, 0.83)] (Supplement Figure 3B).

Sensitivity analysis

Sensitivity analysis was performed to test the stability of the results by excluding each study successively. The results were not affected by sequential exclusion of any particular trial, except for one study (Bains et al., 2016). The detailed sensitivity analysis results are depicted in Fig. 6.

Fig. 6
figure 6

a Sensitivity analysis post-diagnosis aspirin use and overall survival for colorectal cancer. b Sensitivity analysis post-diagnosis aspirin use and cancer specific survival for colorectal cancer. c Sensitivity analysis both pre and post-diagnosis aspirin use and overall survival for colorectal cancer

Full size image

Publication bias

In a meta-analysis with few studies (less than 10), the power of asymmetrical tests is too low to distinguish chance from real asymmetry. Because of the limited number of included studies, it was difficult to confirm the existence of publication bias in the current meta-analysis.

Discussion

Aspirin is a nonselective cyclooxygenase inhibitor. Many studies [2,3,4,5,6,7] have observed that aspirin can improve the prognosis of digestive malignant tumors. However, there were some controversial issues in these studies, especially among those studies that focused on esophageal, gastric, and colorectal cancers with different gene mutation types, such as PIK3CA, that have survival benefits. This meta-analysis included 17 recent clinical studies with large sample sizes to investigate the effects of aspirin on the long-term survival of esophageal, gastric and colorectal cancers. Although the studies included were retrospective studies, they were of high quality and had large sample sizes. The results indicated that postdiagnosis aspirin use may improve OS and CSS in patients with colorectal cancer but not in patients with esophageal cancer or gastric cancer. Subgroup analysis indicated that postdiagnosis aspirin use could prolong the long-term survival of patients with PIK3CA gene mutations and high expression of PTGS2 (COX-2).

A Dutch cohort study [7] that involved 946 patients with esophageal cancer and 750 patients with gastric cancer demonstrated that postdiagnosis aspirin use significantly reduced mortality in esophageal cancer [HR = 0.42, 95% CI (0.30–0.57)] but failed to observe reduced mortality in gastric cancer [HR = 0.87, 95% CI 0.47–1.61]. Additionally, a British study [11] that included4654 patients with esophageal cancer and 3833 patients with gastric cancer observed that low-dose aspirin use did not reduce mortality in these patients. The present study also found that aspirin did not improve the overall survival rate of patients with esophageal and gastric cancer. Although the original studies had high quality and large sample sizes, more RCTs and evidence-based studies are needed because there are few studies that have focused on the long-term survival of patients with esophageal or gastric cancer.

Previous prospective studies [25, 26] have observed that aspirin can reduce colorectal adenomas and reduce the risk of colorectal adenomas recurrence. Most studies have found that aspirin should be used at least one year. The optimal dosage and duration is not consistent and large-scale prospective studies are still needed. This meta-analysis further supports that postdiagnosis aspirin use can improve the long-term survival of patients with colorectal cancer; however, prediagnosis aspirin use cannot improve the long-term survival of patients with colorectal cancer. As aspirin can lead to gastrointestinal bleeding and other side effects, it remains unclear whether low-dose aspirin can achieve adequate antitumor effects. Therefore, the long-term survival of patients with colorectal cancer needs to be evaluated with aspirin in the optimal dose and the best course of treatment. Moreover, side effects on the survival benefit of patients need to be investigated in the future. The daily dose of aspirin in the included observational studies was 75 mg–325 mg, and studies [27, 28] have shown that 81 mg aspirin is sufficient to inhibit rectal mucosal PGE2 production. The US Preventive Services Working Group [29] recommends 81 mg as a prescription dose for aspirin for the primary prevention of cardiovascular diseases and colorectal cancer. However, due to data limitations, a dose-response analysis between aspirin use and the long-term survival of patients with colorectal cancer was difficult to ascertain in the present study, and the optimal course of aspirin treatment needs to be investigated. We perform stratified analysis according to tumor stages. In patients with I-IV, aspirin may increase the overall survival (HR [0.88 (0.79, 0.98)]) and cancer-specific survival (HR [0.85 (0.74, 0.98)]) as shown in supplement Figure 3. We found that aspirin may increase CSS HR [0.73 (0.63, 0.85)] in stage II patients, but there was no survival benefit in other stages. Due to the limited literatures and high heterogeneity, more literatures need to be included for further analysis. Because the inclusion studies did not provide detailed information, it was impossible to conduct a subgroup analysis based on whether surgery or chemotherapy.

The mechanism of action of aspirin in the treatment of colorectal cancer is unclear. Some biomarkers can be used to predict the survival benefit of aspirin in colorectal cancer, including PTGS2 (COX-2) expression and the effects of the PIK3CA gene. The anti-inflammatory effects of aspirin are mediated through direct inhibition of COX-1 and COX- 2 [30,31,32]. PTGS2 (COX-2) promotes the inflammatory response and cell proliferation, and high expression of PTGS2 (COX-2) is associated with poor survival in patients with colorectal cancer [33, 34]. The up-regulation of PI3K enhances PTGS2 (COX-2) activity and prostaglandin synthesis and plays an important role in the signal transduction pathway of tumorigenesis [35, 36]. According to the subgroup analysis in our study, the effects of aspirin use on PIK3CA gene mutation and survival of patients with high expression of PTGS2 (COX-2) was different from that of patients with wildtype PIK3CA and PTGS2 (COX-2)-negative colorectal cancer. These findings provide a basis for the use of aspirin in patients with different types of mutations in colorectal cancer and the result can be used as a preliminary basis for further research.

Due to the bias of retrospective articles, it is necessary to perform randomised prospective studies to validate these data. At present, many clinical trials about aspirin and GI malignancies have not been completed. The ASAC trial (NCT03326791) are the first clinical interventional trial to assess the beneficial role of ASA in recurrence of CRC liver metastases and survival. Add-Aspirin (NCT02804815) aims to assess whether regular aspirin use after standard curative therapy can prevent recurrence and improve survival in individuals with non-metastatic common tumours. ASPIK French trial (NCT02945033) investigate Aspirin Versus Placebo in Resected Colon Cancer With PI3K Mutation Stage III or II High Risk. We also look forward to more prospective studies supporting the impact of aspirin on the prognosis of GI malignancies.

There were some limitations in this study. First, because the original studies were retrospective, there was some publication bias and selection bias. Second, due to the different definitions of aspirin use in the literature, the inclusion and exclusion criteria of the original studies were inconsistent; such differences may lead to deviations in the results. In addition, the number of studies involved was relatively small. Other potential confounding factors include the staging of tumors, whether surgery was performed, whether chemotherapy was performed, and the location of colorectal tumors. Because the included studies did not provide detailed information, it was impossible to conduct a subgroup analysis according to whether surgery, whether chemotherapy, the dosage, duration and reason for taking aspirin.

Conclusion

In conclusion, based on the results of this study, aspirin can improve OS and CSS in patients with colorectal cancer after diagnosis, especially in those with PIK3CA gene mutations and high PTGS2 (COX-2) gene expression, but it cannot improve OS in patients with esophageal cancer and gastric cancer. The results provide a theoretical basis for the conductance of future RCTs. If RCTs can further confirm that aspirin can improve the long-term survival of patients with colorectal cancer, such therapies will have important clinical significance and socioeconomic value for patients with colorectal cancer because aspirin is inexpensive.