Association of Periodontal Red Complex Bacteria With the Incidence of Gastrointestinal Cancers: A Systematic Review and Meta-Analysis

Porphyromonas gingivalis is the primary microbe in the “periodontal red complex” bacteria (PRCB) along with Tannerella forsythia and Treponema denticola, which are linked to periodontal disease (PD). These pathogens are also implicated in various systemic disorders, but their association with the incidence of gastrointestinal (GI) cancer is less explored. A systematic review followed by a meta-analysis was conducted as per standard guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) 2022) to find this association between GI cancers and PRCB after a literature search for full-text papers in the English language (between 2010 and 2023) in databases (Cochrane Library, PubMed, and Web of Science) with suitable keywords using the Boolean search strategy. Data extraction involved titles, abstracts, and full texts retrieved and scored by the modified Newcastle-Ottawa Scale. The data were analyzed by the Review Manager (RevMan 5.2, Cochrane Collaboration, Denmark). Standard Cochran Q test and I2 statistics (for heterogeneity) and a random effects model (pooled OR with 95% CI) were applied to report results. P. gingivalis among the PRCB was linked to GI cancers (OR: 2.16; 95% CI: 1.34-3.47). T. forsythia and T. denticola did not show meaningful associations as per existing evidence for GI cancers.


Introduction And Background
Gastrointestinal (GI) cancers represent the varied types of cancers occurring in the digestive system, which include esophageal, gastric, colorectal, pancreatic, and hepatocellular cancers [1].The global incidence is alarming, with 35% of all deaths caused by cancers.The incidence of GI cancers accounted for 26% of all types of cancers worldwide.The frequency of GI cancers is double in men compared to women.It was reported that the distribution of certain variants of GI cancers is explicit in different regions of the world.Gastric, liver, and esophageal variants were common in Asia [2].Alcohol and tobacco consumption, poor lifestyle (and related disorders such as obesity and overweight), Helicobacter pylori infection, and gut dysbiosis are known risk factors for GI cancers [3].It was reported that the global mortality rate due to GI cancers in 2008 was 39.4 cases per 100,000 individuals, while the incidence rate was 54.9 cases per 100,000.Colorectal and stomach cancers were ranked as the third and fourth most common cancers, respectively.The highest mortality rate was observed among those with stomach cancer (26.1%) of all GI cancers [4].
Oral dysbiosis, or altered oral microbiota, was recently linked to have a significantly positive association with GI cancers (esophageal, pancreatic, and colorectal cancer (CRC)).Dysbiosis of the oral microbiota is caused by two major oral pathologies: dental caries and periodontitis or periodontal disease (PD).Increased colonization of subgingival gram-negative anaerobic bacteria results in deep pockets, loss of clinical attachment, and bleeding on probing (BOP), all corresponding to PD [5,6].The advancement of PD leads to a transition of oral microorganisms toward the "periodontal red complex" bacteria (PRCB) (Porphyromonas gingivalis, Tannerella forsythia, and Treponema denticola), which are known to contribute to the carcinogenesis of GI cancers.Altered oral microbes with periodontal pathogens like T. forsythia, Streptococcus anginosus, Prevotella intermedia, and P. ginigivalis demonstrated a high risk of esophageal adenocarcinoma (EAC).Fusobacterium nucleatum and T. denticola were linked to intensifying the risk of CRC, while P. gingivalis contributed to the pathogenesis of pancreatic cancer (PC) [7].Overall, PRCB was concomitant with the risk of developing different types of GI cancers [8].PRCB presence is linked to many other systemic diseases or conditions, i.e., from liver cirrhosis to chronic kidney disease [9,10].P. gingivalis was reported to promote the occurrence and to worsen the prognosis of esophageal cancer (EC) [11].T. forsythia and P. gingivalis are known to induce a systemic inflammatory response, TNF-alpha pathway activation, and in the generation of reactive oxygen species (ROS) as a part of their GI carcinogenesis [12].Oral microbes inhibit the host immune response and alter the regular inflammatory signaling pathways, leading to an eventful longstanding chronic PD.Periodontal pathogens are also identified as risk factors for atherosclerosis, diabetes, chronic obstructive pulmonary disease, and rheumatoid arthritis, all of which seem to have chronic inflammation and immune alteration in their basic pathogenesis [7,13].
The oral cavity harbors an intricate ecosystem of microbiota, playing a crucial role in pathogen resistance, homeostasis maintenance, and immune system modulation.Growing evidence points to the potential implication of oral microbiota in inducing and promoting GI cancers.This study aimed to identify the role of PRCB in the incidence of various GI cancers, which is a topic less studied in dental literature.

Research Question
The review was conducted as per the standards of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines [14].The research question (as per Participants, Exposure, Control, and Outcomes (PECO)) principle was the following: "Is PRCB infection associated with GI cancers?" P (Participants): subjects with periodontitis associated with one or more PRCB; E (Exposure): PRCB; C (Control): controls (subjects without PD); and O (Outcome): incidence of reported GI cancers.

Selection Process
The prospective or retrospective studies that compared patients with PRCB infection vs. PRCB uninfected cases of GI cancer, studies that had standards for the diagnosis of cancer (confirmed by histopathological examination) and PD (diagnosed by clinical diagnostic criteria), those in line with the current research question, those studies carried out in adults of either gender who were 18 years of age (or above), and those studies that analyzed the cancer incidence rates were eligible.Studies that defined GI cancers/PD without a standard clinical diagnosis did not measure defined outcomes, were published in languages other than English, or were narrative reviews or animal studies were omitted.

Search Strategy
The search strategy involved the use of "keywords or near matches" (including the Medical Subject Headings (MeSH) terms) separated by the bullion operator (Boolean search strategy) "AND or "OR" as per the need.

Data Extraction
Authors (A7, A8, and A9) independently extracted meta-data about the year of publication, isolation method, names of PRCB, sample size, type of GI cancer, and any other significant findings.The studies with relevant titles were collected as per the search strategy, and abstracts were read manually by these authors, if found suitable with defined eligibility, the full texts were obtained and read.The enrolled studies were checked for their value as per the PRISMA checklists by one author (A10).

Data Items
The data of the authors, year of publication, isolated PRCB species, the organs affected under GI cancers, the sample size of cases, and the controls were metadata retrieved after screening full texts of eligible studies.Also, the diagnostic methods used and principal outcomes reported from each study were retrieved along with the said meta-data of selected studies.

Effect Measures
The data were collected and analyzed by Review Manager 5. pooled OR and 95% CI were calculated by a random effects model in these cases.A fixed effects model was applied generally when the heterogeneity was not significantly detected.A mean sample size was taken as a boundary when studies had gross differences or small and large sample sizes.

Quality and Reporting Risk of Bias
The modified Newcastle-Ottawa Scale was used for scoring the quality of included studies, keeping in mind factors such as selection, comparability, and outcomes [15].The studies included after final scoring are as per the final review by a team of four authors (A5-A9).Also, two reviewers analyzed the risk of bias (ROB) using the Egger and Begg regression intercept tests.The STATA 12 (StataCorp LLC, College Station, Texas, USA), wherein a P < 0.05 (two-tailed), was taken for significance.

Study Characteristics and Results of Syntheses
In total, we had a hit of 1,886 search findings from all the mentioned databases within a set timeframe.No further leads were obtained from other sources, such as unpublished data or pilot study results, after verification with authors via personal emails or social platforms (ResearchGate).Of these, 602 were obtained after removal of duplicates, and 204 after excluding (n = 398) reading the abstracts.The authors deemed 68 full papers eligible for review after excluding 60.Finally, eight studies [16][17][18][19][20][21][22][23] were considered for qualitative and quantitative analyses.The process and reasons for elimination at each step are shown in the PRISMA flowchart (Figure 1).Most of the studies had a good sample size (range: 80-821).Of the eight studies, the majority, i.e., Peters et al. [16], Chang et al., and Yuan et al. [23] have studies have analyzed PRCB mostly from patients with EC; two studies each studied PC and CRC.One study by Sun et al. [19] studied precancerous lesions, which was also taken into account here.For the isolation of PRCB samples, 16S rRNA gene sequences from mouthwash were a common method, followed by the use of saliva/blood-based detection by quantitative polymerase chain reaction (qPCR)/immunoblot assays.Kerdreux et al. [22], in cases of CRC, studied the mucosal and fecal samples, while Chang et al. [21] studied cancer tissue specimens directly for the bacteria.All the authors have explored links between GI cancer and P. gingivalis.T. forsythia was next commonly researched, while T. denticola was least explored.Peters et al. [16], Sun et al. [19], and Yang et al. [20] have studied all PRCB and other periodontal pathogens.
P. gingivalis was linked to all types of GI cancers, namely, esophageal squamous cell carcinoma (ESCC)/EAC (EC), PC, and CRC.T. forsythia was linked to EAC/EC but not studied or found linked to PC. T. denticola was the least studied PRCB, but as per one study, it was reported to be associated with poor hygiene practices in CRC.For precancers or gastric precancerous lesions (GPCL), surprisingly, P. gingivalis was found to be associated, but T. forsythia and T. denticola were significantly found to be associated.The details of each study and their quality assessments are shown in  [20]   2019 80) [23]   2017

TABLE 2: Quality assessments as per the modified Newcastle-Ottawa Scale for selected studies
Newcastle-Ottawa Scale quality instrument is scored by awarding a point for each answer that is marked with an asterisk (*).The maximum total score is nine points, with 0-2 points indicating poor quality, 3-5 points indicating fair quality, and 6-9 points indicating good quality.The scores are given in column 3 as a, b, c, etc., corresponding to the respective criteria listed in column 2. Most of the included studies had low bias or were under the "fair quality" cadre.
The findings demonstrated that individuals who had an infection with P. gingivalis faced a 2.16 times greater GI cancer risk as opposed to those who were uninfected (OR = 22.02 (95% CI: 1.34-3.12);P = 0.001).T. denticola and T. forsythia did not have significant associations (Table 3).

FIGURE 5: Assessments of risk of bias for studies/publications
The Begg and Egger regression intercept test indicated no potential publication bias for the incidence of GI cancers (0.091).A two-tailed *P < 0.05 was considered statistically significant.

Discussion
The relationship between tumorigenesis and the activation of chronic inflammation is well-established [24].
H. pylori being associated with pathogenies of GI cancer is direct evidence of the fact that microbial infection can alter the inflammatory signaling pathway and aid in the propagation of carcinogenesis.The absence of timely intervention in such a case may lead to increased spread and, thus, poor prognosis for GI cancers [25].
Researchers have taken an interest in the linking between inflammation (caused by microbiota in PD) and the impact of the same on the onset and/or advancement of systemic diseases.These were proven true for many conditions ranging from Alzheimer's disease to cancers [26].P. gingivalis of the PRCB is reported to promote oral cancers via ATR and NLRP3 inflammasome regulation or γδ T cell activation [27,28].However, the association between these PRCB and the incidence of GI cancers needs to be studied as the oral cavity is an extension of the GI system or alimentary canal.
The current systematic review included eight studies that assessed the associated risk of one or all of the PRCB with the incidence of different types of GI cancers.The eight studies included in the review encompassed a total of 3,400 patients.Studies in recent decades have established a significant association between dysbiosis of the oral microbiota, especially colonization of red complex bacteria, and the incidence of GI cancers.
In the past, GI cancers were linked to smoking, alcohol consumption, obesity, and high cholesterol, all of which fall under the category of modifiable risk factors [29].The more recent data from the past two to three decades have shown growing evidence that oral microbiota plays a significant role in GI cancers.Oral microbes possess a potential ability to trigger carcinogenic metabolism through various pathways in those who have traditional risk factors like smoking or alcohol consumption.This may be achieved by direct conversion of alcohol-/smoking-related carcinogens via oral microorganisms in both oral and GI cancers [30].Oral microbiota with higher levels of subgingival periopathogens was found in patients with digestive system cancers.A study showed that gingival crevicular fluid (GCF) has a higher concentration of T. forsythia, P. gingivalis, and T. denticola in patients with gastric cancer diagnosed with PD [31].
P. gingivalis is one of the most studied PRCB and is shown to be concomitant in the pathogenesis of GI cancers.Tumorigenic markers like colony-stimulating factor 1 (CSF1), colon cancer-associated transcript 1 (CCAT1), and cellular invasion markers like MMP-9 were enhanced in P. gingivalis infected cells when patients with GI cancers and, more specifically, for CRC [32].P. gingivalis was found to be positively associated with ESCC (also P. gingivalis IgG and IgA) and also increases the risk of PC by three times [33,34].Available literature has elucidated the correlation between P. gingivalis and the development of ESCC through various methods.The salivary samples of the patients with ESCC were significantly high in P. gingivalis when compared to healthy individuals, as estimated using 16S rRNA gene sequencing.P. gingivalis is reported to promote the proliferation and metastasis of ESCC through NF-kB pathway activation [35].
Another study that analyzed tissue specimens from 100 ESCC patients detected P. gingivalis and Kgp enzymes in 61% of the samples through immunohistochemistry and qPCR [36].These significant pieces of evidence suggest that P. gingivalis is associated with the development and progression of ESCC, aligning with the findings of the present review.
P. gingivalis was reported to be implicated in the tumorigenesis of PC by stimulating the genes (such as s100a8 or cxcl1/2) associated with neutrophil chemotaxis.This can be considered as a direct epigenetic mechanism in GI and pancreatic carcinogenesis [37].A murine model demonstrated a higher genetic expression of Reg3g after stimulating the pancreas with P. gingivalis lipopolysaccharide, pointing to a possible association with PC [38].An in vivo study showed that P. gingivalis can adhere, invade, and promote cell proliferation in host cells through MAPK and ERK pathways of colon cancer or CRC.It was also established that abundant P. gingivalis was observed in CRC tissue and fecal samples [39,40].
Mediators of inflammation (such as interleukin 1 and tumor necrosis factor) common to inflammation and carcinogenesis are triggered by T. forsythia to be produced from macrophages.T. forsythia was associated with EAC [41].Likewise, the presence of T. denticola chymotrypsin-like protease (Td-CTLP) in various GI cancer specimens is documented.The IHC study demonstrated the presence of Td-CTLP in oropharyngeal SCC and other types of orodigestive cancer specimens (tongue, tonsils, and esophagus) [42,43].These findings did not elucidate a direct association between T. denticola and the development of GI cancers, which needs further research.Yet another prospective study that explored the antibodies of PRCB about the incidence of cancer in 1,712 years of follow-up showed that T. denticola antibodies exhibited an inverse relationship to colon cancer (HR = 1.52) [44].Put together, the existing data proposes that P. ginigivalis has a greater role in the development of GI cancers than the other two PRCBs, which is in line with the findings of the current study.The role of T. denticola and T. forsythia in the incidence of GI cancers requires further probing to determine proper incidences or associations, followed by understanding their molecular mechanisms.Also, we found that most studies done were on P. gingivalis only and not the other PRCB.
The limitations of the study lie in the selection of recent evidence only, which may have underrepresented papers on T. denticola.Similarly, there was substantial overall heterogeneity, necessitating the use of random effects models for the analysis.Another limitation of the study is the differences in the medium from which PRCB was extracted.Mouth rinse samples (studies on patients with PC, OC, or CRC) followed by blood, saliva, feces, and direct isolation from tissue were done in these studies.Each medium used is a known standard for studying that specific GI cancer (e.g., fecal samples for CRC), which led us to include the studies.However, the variations due to these differing samples used for PRCB extraction were not evaluated for their effect on the study outcome.The future directions may include studying more periodontal bacteria from populations of different ethnicities and adjusting for all described confounders with implications for GI cancers.

Conclusions
The study showed that individuals infected with P. gingivalis have a raised risk of GI cancer by 2.16 times as opposed to those who are uninfected.T. denticola and T. forsythia did not show significant such associations.

FIGURE 1 :
FIGURE 1: PRISMA flow diagram GI, gastrointestinal; PRCB, periodontal red-complex bacteria; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analysis (1.76(1.19-2.60))was associated with an increased risk of CRC.No associations with T. forsythia and P. gingivalis were reported.was detected in 60.7% of OSCC tissues as opposed to 13.3% of normal tissues.P. gingivalis infection was linked to poor prognosis in patients with OSCC who had PD progression (tooth loss, deep pockets, or higher rates of was detected significantly in the feces of CRC patients compared to controls (P = 0.028).A positive association was found between the presence of P. gingivalis in feces and tumor tissue (P < 0.001).

FIGURE 2 :
FIGURE 2: Forest plot showing the association of P. gingivalis with the incidence of gastrointestinal (GI) cancersForest plot showing individual OR (black boxes) taken from selected studies (on the y-axis) and range (x-axis) for associating P. gingivalis with the incidence of GI cancers.The sum effect or pooled OR derived was OR; 95%CI, P) = [1.86(1.20,2.88)P = 0.001] (white diamond mark).

FIGURE 3 :
FIGURE 3: Forest plot showing the association of T. forsythia with the incidence of gastrointestinal (GI) cancersForest plot showing individual OR (black boxes) taken from selected studies (on the y-axis) and range (x-axis) for associating T. forsythia with the incidence of GI cancers.The sum effect or pooled OR derived was OR; 95% CI, P) = [1.06(95% CI, 0.8-1.44)P = 1.0] (white diamond mark).

FIGURE 4 :
FIGURE 4: Forest plot showing the association of T. denticola with the incidence of gastrointestinal (GI) cancersForest plot showing individual OR (black boxes) taken from selected studies (on the y-axis) and range (x-axis) for associating T. denticola with the incidence of GI cancers.The sum effect or pooled OR derived was [OR; 95%CI, P) = [1.31(0.97-1.71)P = 0.6] (white diamond mark).

Table 1 ,
while their respective quality assessments are in