Aspirin and Cancer Survival: An Analysis of Molecular Mechanisms

Simple Summary The use of aspirin has shown a definite role in the prevention of cancer; however, its effect on survival is still debated. The evidence from randomized trials failed to show any benefit, while cohort studies have demonstrated its usefulness. This is attributed to the use of different doses of aspirin in differing studies. This article explores the plausible mechanisms through which aspirin may exert its effect on improving survival. It is possible that the use of aspirin as adjunct to standard care may lead to better survival in cancer, though the actual effect would have to be demonstrated in clinical trials. Abstract The benefit of aspirin on cancer survival is debated. Data from randomized clinical trials and cohort studies are discordant, although a meta-analysis shows a clear survival advantage when aspirin is added to the standard of care. However, the mechanism by which aspirin improves cancer survival is not clear. A PubMed search was carried out to identify articles reporting genes and pathways that are associated with aspirin and cancer survival. Gene ontology and pathway enrichment analysis was carried out using web-based tools. Gene–gene and protein–protein interactions were evaluated. Crosstalk between pathways was identified and plotted. Forty-one genes were identified and classified into primary genes (PTGS2 and PTGES2), genes regulating cellular proliferation, interleukin and cytokine genes, and DNA repair genes. The network analysis showed a rich gene–gene and protein–protein interaction between these genes and proteins. Pathway enrichment showed the interleukin and cellular transduction pathways as the main pathways involved in aspirin-related survival, in addition to DNA repair, autophagy, extracellular matrix, and apoptosis pathways. Crosstalk of PTGS2 with EGFR, JAK/AKT, TP53, interleukin/TNFα/NFκB, GSK3B/BRCA/PARP, CXCR/MUC1, and WNT/CTNNB pathways was identified. The results of the present study demonstrate that aspirin improves cancer survival by the interplay of 41 genes through a complex mechanism. PTGS2 is the primary target of aspirin and impacts cancer survival through six primary pathways: the interleukin pathway, extracellular matrix pathway, signal transduction pathway, apoptosis pathway, autophagy pathway, and DNA repair pathway.


Introduction
Despite accumulating evidence of the benefit of aspirin against cancer, its effect on improving cancer survival is still debated, since the mechanism by which it impacts cancer survival is not completely understood and the published data are discordant.There have been four randomized controlled trials (RCTs) [1][2][3][4] showing mixed results from no effect to improved survival.Lipton  or placebo for two years, and demonstrated no difference in disease-free or overall survival rates [3].A second RCT in 1991 randomized patients with renal cancer to interferon and interferon plus 600 mg of aspirin and reported better results with interferon alone [1].
The third randomized study, conducted in 1993, randomized patients with lung cancer to chemotherapy or chemotherapy with aspirin.In this trial, a daily dose of 1000 mg of aspirin was used, because this dose is supposed to influence platelet function.However, this study also failed to show any benefit of adding aspirin to the standard of care [2].All these three trials used a relatively high dose of aspirin, ranging from 600 mg to 1000 mg.However, in 2009, Liu et al., for the first time, reported improved survival in esophageal cancer patients randomized to a very low dose of aspirin (50 mg) [4].Since then, several retrospective and observational studies have reported a survival advantage of adding aspirin to the treatment for various cancers.A meta-analysis of 118 studies, 63 of them specifically reporting on cancer mortality and the rest on all-cause mortality, found a 21% reduction in cancer deaths and about 20% reduction in all-cause mortality (pooled hazard ratio (HR): 0.79; 95% confidence intervals: 0.73, 0.84) [5].
Various mechanisms have been proposed to explain the protective role of aspirin against cancer.The inhibition of COX2, regulating apoptosis, and reduction in angiogenesis through the prostaglandin pathway are predominant; however, aspirin is also said to improve survival by regulating platelet function and reducing metastasis [12,13].Other proposed mechanisms include the inhibition of the peroxisome proliferator-activated receptor (PARP) and hence interference with the homologous recombinant (HR) DNA repair pathway, nuclear factor-kB (NFκB), and PI3KC pathway regulation.However, no bioinformatic studies have evaluated the interaction of genes associated with cancer survival and their protein products to investigate the possible mechanism by which aspirin can affect cancer survival.Therefore, we conducted this systematic review and bioinformatic analysis to explore the possible mechanism(s) by which the addition of aspirin may produce a survival benefit in cancer patients.

Material and Methods
A PubMed search was carried out filtering for English-language papers and using the string (("aspirin"[MeSH Terms] OR "aspirin"[All Fields] OR "aspirins"[All Fields] OR "aspirin s"[All Fields] AND ("cancer s"[All Fields] OR "neoplasms"[MeSH Terms] OR "neoplasms"[All Fields] OR "cancer"[All Fields] OR "cancers"[All Fields]) AND ("mortality"[MeSH Subheading] OR "mortality"[All Fields] OR "survival"[All Fields] OR "survival"[MeSH Terms] OR "survivability"[All Fields] OR "survivable"[All Fields] OR "survivals"[All Fields] OR "survive"[All Fields] OR "survived"[All Fields] OR "survives"[All Fields] OR "surviving"[All Fields]) AND ("genes"[MeSH Terms] OR "genes"[All Fields] OR "gene"[All Fields] OR ("gene s"[All Fields] OR "genes"[MeSH Terms] OR "genes"[All Fields]) OR ("genome"[MeSH Terms] OR "genome"[All Fields] OR "genomes"[All Fields] OR "genome s"[All Fields] OR "genomically"[All Fields] OR "genomics"[MeSH Terms] OR "genomics"[All Fields] OR "genomic"[All Fields])) to identify articles reporting on genes and pathways that are associated with aspirin and cancer survival.Studies reporting on the molecular mechanism either in clinical or experimental studies on cell lines were evaluated.The search was restricted to English.Non-English studies and studies reporting on NSAID's other than aspirin or aspirin in combination with other drugs were excluded.The genes were identified, and bioinformatic analysis was performed.
WEB-based Gene Set Analysis Toolkit (Webgestalt) (http://www.webgestalt.org/)was used to perform gene ontology (GO) and pathway enrichment analysis, and GeneMANIA (https://genemania.org)was used for gene-gene interactions.A protein-protein interaction network (PPI) was constructed using NetworkAnalyst (http://www.networkanalyst.ca), and interacting genes were searched using Search Tool for the Retrieval of Interacting Genes (STRING) (http://string-db.org/).Pathways were created using Reactome (https://reactome.org/PathwayBrowser/), and the rest of the data were manually curated using public databases.Crosstalk between enriched pathways was built using the ShinyGO v0.741 tool (http://bioinformatics.sdstate.edu/go74/).Based on a false discovery rate (FDR) p value of <0.05, the top 30 pathways were enriched.Using the p values from these 30 pathways, a hierarchical clustering tree was constructed.Crosstalk between the genes was prepared manually from the network.All the above websites were accessed on 20 August 2022.

Figure 1. (A)
Gene-gene interaction of the 11 main genes that explain how aspirin can improve cancer survival (grey shaded nodes are primary input genes, grey non-shaded nodes are the secondary genes showing interaction with primary genes) (B).Protein-protein interaction of 11 proteins synthesized by the 11 main genes identified, showing three clusters of proteins; the first cluster is co-expression of PTGS2, PTGES2, and P53 (red), while the second cluster is of cell cycle regulators (green), and the third is of DNA repair genes (blue).
Given this confirmation that PTGS2 and PTGES2 are the primary genes inhibited by aspirin, we evaluated how these two primary genes interact with the other 39 genes, which we categorized into three groups based on their function, i.e., cell signaling and cell proliferation, cytokines and interleukins, and tumor suppressor genes.The analysis of PTGS2 and PTGES2 with genes participating in cell signaling and proliferation showed 37% co-expression and 20% physical interaction with 18 primary nodes and 20 secondary nodes, reflecting how these genes relate to one another (Figure 2A).The protein-protein network of these three groups of genes had 18 nodes and 62 edges (Figure 2B).The p value of the network was highly significant (p = 2.75 × 10 −15 ), suggesting significantly more interactions than expected.The highest number of interactions with PTGS2 was seen with the MAP kinase and EGFR pathways, suggesting that COX2 (PTGS2) suppression may lead to inhibition of MAP kinase and EGFR pathways and hence a reduction in cellular growth and proliferation.
We then examined the interaction of PTGS2 and PTGES2 with the interleukin and cytokine genes.The gene-gene network showed a high level of co-expression of these genes (85%), which was expected as the COX2 gene, and prostaglandin controls the synthesis of interleukins and cytokines (Supplementary Figure S1A).The protein-protein network of these genes was equally rich, with 18 nodes and 122 edges (Supplementary Figure S1B).The PPI enrichment value was highly statistically significant (p ≤ 1.0 × 10 −16 ), indicating that there were many more interactions than expected by chance. of these three groups of genes had 18 nodes and 62 edges (Figure 2B).The p value of the network was highly significant (p = 2.75 × 10 −15 ), suggesting significantly more interactions than expected.The highest number of interactions with PTGS2 was seen with the MAP kinase and EGFR pathways, suggesting that COX2 (PTGS2) suppression may lead to inhibition of MAP kinase and EGFR pathways and hence a reduction in cellular growth and proliferation.We then examined the interaction of PTGS2 and PTGES2 with the interleukin and cytokine genes.The gene-gene network showed a high level of co-expression of these genes (85%), which was expected as the COX2 gene, and prostaglandin controls the synthesis of interleukins and cytokines (Supplementary Figure S1A).The protein-protein network of these genes was equally rich, with 18 nodes and 122 edges (Supplementary Figure S1B).The PPI enrichment value was highly statistically significant (p ≤ 1.0 × 10 −16 ), indicating that there were many more interactions than expected by chance.The network of PTGS2 and PTGES2 with tumor suppressor genes showed nearly 70% physical interactions with 21 secondary nodes between these genes (Supplementary Figure S2A).While the protein-protein interaction network (Supplementary Figure S2B) had five nodes and five edges, the interaction was not statistically significant (p = 0.07), suggesting that COX2 does not directly regulate these tumor suppressor genes.In contrast, the cell signaling and proliferation proteins (Supplementary Figure S2C) and the cytokine and interleukin gene proteins (Supplementary Figure S2D) did interact with the tumor suppressor gene proteins (p ≤ 1.0 × 10 −16 ), suggesting that aspirin indirectly regulates tumor suppressor genes through its impact on cell signaling and the proliferation gene and interleukin and cytokine gene expression.All gene-gene interactions between the four groups (primary genes, cell signaling and cell proliferation genes, cytokine and interleukin genes, and tumor suppressor genes) are detailed in Supplementary File S2.
A pathway analysis with Reactome showed that the largest number of genes interacted with interleukin (Supplementary File S3) and signal transduction pathways (Supplementary File S4).Additional interactions were identified in DNA repair, autophagy, extracellular matrix, and apoptosis pathways (Supplementary Files S5-S8).Based on these pathways, a pathway diagram was created that incorporated most genes and pathways (Figure 3A,B).The crosstalk between the genes and the pathways is shown in Figure 4.
kin genes, and tumor suppressor genes) are detailed in Supplementary File S2.
A pathway analysis with Reactome showed that the largest number of genes interacted with interleukin (Supplementary File S3) and signal transduction pathways (Supplementary File S4).Additional interactions were identified in DNA repair, autophagy, extracellular matrix, and apoptosis pathways (Supplementary Files S5-S8).Based on these pathways, a pathway diagram was created that incorporated most genes and pathways (Figure 3A,B).The crosstalk between the genes and the pathways is shown in Figure 4.

Discussion
There are many studies on cancer prevention using aspirin, and its effect, along with that of other COX2 inhibitors, is well known.The results of the randomized controlled trials have been mixed, and the findings from the low-dose trial and subsequent non-randomized observational studies suggest that the addition of aspirin to the standard of care

Discussion
There are many studies on cancer prevention using aspirin, and its effect, along with that of other COX2 inhibitors, is well known.The results of the randomized controlled trials have been mixed, and the findings from the low-dose trial and subsequent non-randomized observational studies suggest that the addition of aspirin to the standard of care improves cancer survival.However, the mechanism by which aspirin improves survival is not clear.This study is the first to our knowledge to comprehensively evaluate the literature and apply bioinformatics to investigate the underlying mechanism.
Our literature search identified 41 genes that are related to aspirin and cancer survival and that fell into four clusters, including the primary genes directly regulated by PTGS2 and PTGES2, oncogenes and cell cycle regulators (16 genes), interleukin and cytokines regulators (16 genes), and tumor suppressor genes (7 genes).Of the 11 main genes that are regulated by aspirin or interact with aspirin, the DNA repair pathway, especially the homologous recombination (HR) pathway, showed co-expression of PTGS2 with TP53, PARP1, and PARP2.
Several previous studies have shown that aspirin can affect DNA repair of genes.For example, treatment of a DNA MMR competent/p53 mutant colorectal cancer cell line with aspirin for 48 h led to DNA damage pathway gene expression, including BRCA1 [45].A later study found that feeding aspirin to Dalton cell lymphoma-bearing mice resulted in cell cycle arrest in the G0/G1 phase [34].In addition, aspirin was found to lower the number of somatic mutations, including mutations in TP53 in Barret's esophagus patients [40].The potential of altering the levels of NFκB and its use as a biomarker are being tested in ASAMET (NCT03047837) trial, wherein both aspirin and metformin are given to patients with stage I-III colorectal patients, and the results are awaited [46].
Despite the availability of cell line and animal model data, it is still not clear why there is co-expression of COX2 (PTGS2) with DNA repair pathway genes.Rationally, as COX2 increases cellular proliferation, it should inhibit DNA repair and apoptosis proteins, as is seen with counter-regulation of the BCL2/BAX/Caspase cascade.Two possible mechanisms are through the regulation of MYC and MUC1.Alternatively, the co-expression of COX2 and DNA repair pathway genes could be due to an increase in DNA synthesis, thereby increasing the DNA repair cascade to check the DNA that is being newly synthesized.We demonstrate the regulation of DNA repair genes through crosstalk of the interleukin pathway that regulates JAK1, PI3KC, and AKT, which in turn regulate the p53 pathway and another interaction through CXCR regulation through Muc1 (Figure 4).This hypothesis of complex crosstalk might be more plausible, but the exact mechanisms involved are not clearly understood.Additional studies are required to understand this process fully.
The effect of PTGS2/PTGES2 on interleukin pathway activation is more obvious and can be explained by its effect on arachidonic acid metabolism and increased synthesis of interleukins and cytokines.Similarly, the effects on cellular proliferation and signal transduction pathways are also explainable through direct [45] or interleukin mechanisms [23,33,34,47].
Though most of the RCTs failed to demonstrate any survival benefit from the addition of aspirin to the standard of care, the benefit is seen in one RCT of low-dose aspirin, in observational studies, and in a meta-analysis that included these observational studies.This inconsistency could be due to the use of different doses of aspirin or inherent biases in the observational studies like adherence bias, healthy user bias, etc. Further, most earlier studies were based on the effect of aspirin on the blockade of the COX2 or thromboxane (Tx) pathway and assumed this blockade to be the main mechanism through which aspirin exerts its benefit on cancer survival.It is well known that at a low dose, aspirin irreversibly acetylates serine 530 of cyclooxygenase (COX)-1.This inhibits platelets from generating thromboxane A2, thus resulting in an antithrombotic effect by making TBX2A unavailable to bind TBXA2R.However, how this effect reduces cancer survival is not known.The present study is the first to demonstrate that aspirin can improve survival by an interplay of 41 genes, 2 of which (PTGS2 (COX2) and PTGES2) are the primary target of aspirin.

Figure 2 .
Figure 2. (A).Showing gene-gene interactions of PTGS2/PTGES2 with cell cycle regulators and cellular proliferation genes identified by data and text mining (purple-co-expression, pink-physical interaction, green-genetic interaction, orange-predicted interactions) (B).The protein-protein network of PTGS2/PTGES2 with cell cycle regulators and cellular proliferation proteins has 18 nodes and 62 edges.The interactions are statistically significant (p = 2.75 × 10 −15 ) (light blue-curated databases, pink-experimentally determined; green-predicted gene neighborhood; red-predicted gene fusion; blue-predicted gene co-occurrence).

Figure 2 .
Figure 2. (A).Showing gene-gene interactions of PTGS2/PTGES2 with cell cycle regulators and cellular proliferation genes identified by data and text mining (purple-co-expression, pink-physical interaction, green-genetic interaction, orange-predicted interactions) (B).The protein-protein network of PTGS2/PTGES2 with cell cycle regulators and cellular proliferation proteins has 18 nodes and 62 edges.The interactions are statistically significant (p = 2.75 × 10 −15 ) (light blue-curated databases, pink-experimentally determined; green-predicted gene neighborhood; red-predicted gene fusion; blue-predicted gene co-occurrence).

Figure 3 .
Figure 3. (A) Pathway enrichment with crosstalk between identified pathways.(B) Tree analysis of the pathway based on the p value for each pathway on enrichment.Pathways with many shared genes are clustered together.Bigger dots indicate more significant p-values.

Figure 3 . 12 Figure 4 .
Figure 3. (A) Pathway enrichment with crosstalk between identified pathways.(B) Tree analysis of the pathway based on the p value for each pathway on enrichment.Pathways with many shared genes are clustered together.Bigger dots indicate more significant p-values.Cancers 2024, 16, x FOR PEER REVIEW 12 of 12

Figure 4 .
Figure 4. Proposed mechanism of aspirin in inhibiting cellular proliferation and increasing in autophagy, DNA repair, and apoptosis through complex crosstalk.Genes identified through data mining are in shaded boxes.Primary genes are in green text, oncogenes and cell cycle regulator genes in blue text, tumor suppressor genes in red text, cytokines and interleukins in purple text, and secondary genes are in black text.Dotted line show predicted interaction.

Table 1 .
The genes identified as effectors to increase survival in patients receiving Aspirin, their grouping according to the function.
KRAS-Kirsten rat sarcoma viral oncogene homologEncodes tumor oncogene Kras protein part of RAS/MAPK pathway among other signal transduction pathways
Hematopoietic growth factor secreted by stromal cells in the bone marrow and thymus.Participates in proliferation during certain stages of B-cell maturation, T and NK cell survival, development, and homeostasisIL8-Interleukin 8Chemokine produced by macrophages and epithelial cells.Induces chemotaxis and angiogenesis