Association of the Streptococcus bovis/Streptococcus equinus Complex With Colorectal Neoplasia: A Systematic Review and Meta-analysis

Abstract Background Invasive infection with Streptococcus bovis/Streptococcus equinus complex (SBSEC) bacteria is associated with underlying colorectal neoplasia. However, the link between intestinal or fecal colonization with SBSEC isolates or antibody responses to SBSEC members and colorectal cancer is not thoroughly investigated in the literature. Methods We searched the PubMed, EMBASE, and Web of Science databases for case–control studies as well as retrospective or prospective cohort studies reporting an association between SBSEC bacteria and colorectal neoplasia. Results We identified 22 studies (15 case–control and 7 cohort) that met our inclusion criteria. Among the cohort studies, patients with SBSEC bacteremia were 3.73 times more likely to have underlying colorectal cancer compared with individuals with no bacteremia (relative risk [RR], 3.73; 95% CI, 2.79–5.01), whereas the risk of underlying colorectal adenoma in patients with SBSEC bacteremia was not significantly increased (RR, 5.00; 95% CI, 0.83–30.03). In case–control studies, patients with colorectal cancer were 2.27 times more likely to have evidence of intestinal or fecal colonization with SBSEC isolates (odds ratio [OR], 2.27; 95% CI, 1.11–4.62) and immunoglobulin G (IgG) antibody responses to SBSEC antigens (OR, 2.27; 95% CI, 1.06–4.86) compared with controls. Patients with colorectal adenoma were not more likely to be colonized with SBSEC isolates compared with controls (OR, 1.12; 95% CI, 0.55–2.25). Conclusions Apart from the well-established association of SBSEC bacteremia and underlying colorectal cancer, intestinal or fecal colonization with SBSEC isolates and IgG antibody responses to SBSEC antigens were higher in patients with colorectal cancer compared with controls. Neither bacteremia from SBSEC isolates nor colonization with SBSEC bacteria was associated with underlying colorectal adenoma.

Colorectal cancer is the third most common type of cancer and a leading cause of cancer-related mortality on a global scale [1].Alteration in the gut microbiome homeostasis has been recognized as a major contributor to colorectal cancer development [2,3].The Streptococcus bovis/Streptococcus equinus complex (SBSEC) is a group of nonenterococcal group D streptococci that cause invasive disease, mainly bacteremia or infective endocarditis [4].Initially, bacterial isolates belonging to SBSEC were phenotypically designated as S. bovis biotypes I, II/1, and II/2; the advent of molecular classification assays resulted in the reclassification of SBSEC isolates to Streptococcus gallolyticus subsp gallolyticus (I), S. infantiarius subsp infantarius (II/1), S. infantiarius subsp coli (II/1), and S. gallolyticus subsp pasteurianus (II/2) [5].The association of SBSEC bacteremia with underlying colorectal neoplasia has been described in terms of pathogenesis [6,7], and clinical evidence supporting a link between this complex and colorectal cancer is increasing [8], reinforcing the critical role of SBSEC in the detection of occult colorectal malignancy.
Although invasive SBSEC infection can signal the presence of underlying colorectal neoplasia, the detection of SBSEC isolates in asymptomatic patients has not been extensively studied to determine if it aids in the detection of colorectal cancer.The detection of SBSEC isolates in intestinal or fecal tissue samples in the general population ranges from 5% to 60%, with this variability being attributed to diagnostic modalities harnessed to detect SBSEC, and regional differences in terms of distribution of SBSEC bacteria in the population [9,10].In patients with hepatobiliary disorders, the detection rate of SBSEC ranges Open Forum Infectious Diseases widely from 2% to 83%, depending on the SBSEC species under consideration [11][12][13][14][15].In patients with colorectal cancer, colonization rates of up to 74% have been reported [16], with most data linking S. gallolyticus subsp gallolyticus with underlying colorectal malignancy, although associations with other SBSEC bacteria and colorectal cancer have also been described [17,18].The increased incidence of S. gallolyticus subsp gallolyticus colonization in colorectal cancer has been attributed to multiple mechanisms, including preferential attachment of the bacteria to exposed collagen fibers emerging from malignant colorectal tissue [18], increased bacterial competitiveness due to bile acid-induced toxin production that kills competing enterococci in the gastrointestinal tract [19], and utilization of glucose and its metabolites together with amino acids that confer a growth advantage for SBSEC bacteria [20].Moreover, the presence of serum immunoglobulin G (IgG) antibodies against SBSEC antigens has been recently studied as a potential molecular marker for the presence of colorectal neoplasia in asymptomatic individuals.Several studies have demonstrated a positive association between seropositivity against SBSEC antigenic targets and underlying colorectal adenoma or carcinoma [21][22][23][24].
While the link between SBSEC bacteremia and colorectal cancer is well described [16,[25][26][27], the link between intestinal or fecal colonization with SBSEC isolates or serologic responses to SBSEC members and underlying colorectal neoplasia has not been extensively analyzed.In this systematic review and metaanalysis, we evaluated whether fecal or intestinal colonization with SBSEC isolates or evidence of IgG antibody responses to SBSEC antigens was associated with colorectal cancer risk.Furthermore, we aimed to reinforce the link between SBSEC bacteremia and underlying colorectal cancer.

Approach
This systematic review and meta-analysis was performed in accordance with the Meta-Analysis of Observational Studies in Epidemiology (MOOSE) reporting guidelines (Supplementary Table 1) [28] and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement checklist (Supplementary Table 2) [29].

Data Sources
We searched the PubMed/MEDLINE, EMBASE, and Web of Science databases for literature in English (from database inception to June 30, 2023) using the following search terms: (("Streptococcus bovis" OR bovis OR gallolyticus OR infantarius OR pasteurianus OR coli OR lutetiensis) AND (malignancy OR neoplasm OR cancer OR "colorectal cancer" OR "colonic neoplasia" OR "colonic adenocarcinoma" OR carriage OR "fecal carriage")).Manual checking of reference lists of retrieved articles and reports, including relevant papers and metaanalyses, was performed to identify additional and potentially relevant articles.The Rayyan screening tool was used to create our database and screen the exported studies [30].

Study Selection
Observational studies, including case-control and cohort studies, were assessed for inclusion in our analysis.For case-control studies, we defined cases as patients with a biopsy-proven diagnosis of colorectal neoplasia, defined as either colorectal adenoma or carcinoma, while controls were defined as individuals with no diagnosis of colorectal neoplasia.For retrospective or prospective cohort studies, the exposed group included patients aged >18 years with 1 of the following exposures of interest: (a) presence of SBSEC bacteremia and/or infective endocarditis, diagnosed via blood cultures (bacteremia), the Duke criteria, or the modified Duke criteria (infective endocarditis) [31]; (b) evidence of SBSEC colonization, defined as either culture of SBSEC isolates in colonic or fecal tissue samples or identification of SBSEC DNA through detection of the sodA gene via polymerase chain reaction (PCR) or in situ hybridization in colonic or fecal tissue samples; or (c) presence of IgG antibodies against the SBSEC cell wall and pili protein antigens in the serum (detailed information on diagnostic assays, antigenic targets, and cutoff values for IgG antibody positivity are presented in Supplementary Table 3).The control group included individuals with no evidence of SBSEC bacteremia, colonization, or serum IgG antibodies against SBSEC antigens.
The genotypic designation of the SBSEC was used to report associations between SBSEC and colorectal neoplasia, whenever available.

Study Outcomes
For case-control studies, we examined the odds of intestinal or fecal colonization with SBSEC in patients with colorectal cancer or adenoma, as well as serologic responses against S. gallolyticus subsp gallolyticus antigens in patients with colorectal cancer compared with individuals with no diagnosis of colorectal neoplasia.For cohort studies, we examined the risk of colorectal cancer or adenoma in patients with SBSEC bacteremia (with or without concurrent infective endocarditis) compared with patients without bacteremia.

Data Extraction and Quality Assessment
Two reviewers (K.O. and A.G.) independently determined study eligibility by screening titles and abstracts and performing full-text reviews of selected studies.Potential disagreements were resolved by a third reviewer (D.B.).Extractable data included baseline patient characteristics, methods used to diagnose bacteremia or colonization, SBSEC phenotypic and/or genotypic classification, primary and secondary outcomes, and information related to study quality.
Two reviewers (K.O. and A.G.) assessed the quality of the included studies using the Newcastle-Ottawa-Scale (NOS) [32], a tool developed for quality assessment of nonrandomized studies.The NOS checklist is a 9-point scale that involves the appraisal of methodological issues and their reporting.The scoring system encompasses 3 major domains (participant selection, group comparability, and ascertainment of exposure); scores range from 0 to 9, with scores ≥7 indicating high-quality studies.

Data Synthesis and Analysis
For data analysis, we used Stata, version 17 (Stata Corporation, College Station, TX, USA).We performed a random-effects meta-analysis using the restricted maximum likelihood method [33] to estimate the risk of colorectal neoplasia in patients with SBSEC bacteremia compared with patients with no evidence of invasive infection.We also calculated the odds of intestinal or fecal colonization with SBSEC in patients with colorectal cancer or adenoma, as well as serum IgG antibody responses against S. gallolyticus subsp gallolyticus antigens in patients with colorectal cancer compared with individuals with no diagnosis of colorectal neoplasia.We reported odds ratios (ORs) and relative risks (RRs) with 95% confidence intervals.We estimated heterogeneity using the I 2 statistic, with I 2 values of 25%, 50%, and 75% representing low, moderate, and high heterogeneity, respectively [34].We set statistical significance at α = .05.We used the Egger test to evaluate for publication bias and to assess small study effects [35].
We also conducted a subgroup analysis according to the genotypic classification of SBSEC to calculate the risk of colorectal cancer in patients with S. gallolyticus subsp gallolyticus bacteremia compared with individuals without bacteremia, as well as the odds of intestinal or fecal colonization with S. gallolyticus subsp gallolyticus in patients with colorectal cancer compared with controls.

Search Results
We found 5033 studies from our literature search in PubMed, EMBASE, and Web of Science.After the removal of duplicates (n = 1291), we examined 3742 studies for eligibility.We excluded 3715 studies based on title and abstract assessment and found 22 studies that met our eligibility criteria (study schema detailed in Figure 1).

Quality Assessment of the Studies Included in the Analysis
Supplementary Table 4 shows the overall study quality assessment for the case-control studies included in the analysis, and Supplementary Table 5 summarizes the overall study quality assessment for the cohort studies included in the analysis, according to the NOS checklist.The inclusion of studies classified as high quality with regards to participant selection, group comparability, and outcome assessment helps strengthen our

Sensitivity Analysis According to NOS Quality Assessment
We repeated the main analysis by including only the high-quality studies (NOS ≥7) to explore the robustness of our findings and ensure that aspects regarding study design and outcome assessment would not interfere with the interpretation of our meta-analysis results.After excluding fair-quality studies, a subgroup analysis was available only for case-control studies estimating the odds of colonization with SBSEC isolates in patients with colorectal neoplasia compared with controls.A sensitivity analysis was not carried out for evaluating the association between SBSEC bacteremia or IgG antibody responses to SBSEC antigens and colorectal neoplasia due to an insufficient number of high-quality studies.We included 6 high-quality studies reporting information on patients with colorectal cancer [36,38,42,43,45,46], and the sensitivity analysis showed that patients with colorectal cancer were 1.85 times more likely to have evidence of colonization with SBSEC isolates in fecal or intestinal tissues compared with controls (OR, 1.85; 95% CI, 1.01-3.38;I 2 = 0.00%) (Supplementary Figure 3).Next, of the 3 studies that had information about individuals with colorectal adenoma [38,42,43], colonization with SBSEC isolates in fecal or intestinal tissues was not observed, with higher odds in patients with colorectal adenoma compared with controls (OR, 1.01; 95% CI, 0.42-2.46;I 2 = 0.00%) (Supplementary Figure 4).

DISCUSSION
In this systematic review and meta-analysis, we explored the association between SBSEC and colorectal neoplasia.We found that patients with colorectal cancer were 2.27 times more likely to be colonized by SBSEC bacteria or have IgG antibody responses to S. gallolyticus subsp gallolyticus antigens compared with controls.Also, patients with SBSEC bacteremia were 3.73 times more likely to have underlying colorectal cancer compared with controls.Interestingly, our analysis did not find an association between SBSEC bacteremia or colonization and underlying colorectal adenoma diagnosis.
In our analysis, SBSEC bacteremia, and especially S. gallolyticus subsp gallolyticus invasive infection, was associated with underlying colorectal cancer.Consistent with our findings, a previous meta-analysis, including 3 cohort studies with 140 patients with SBSEC bacteremia and 450 controls, concluded that colorectal cancer risk was almost 8 times higher in patients with SBSEC bacteremia [26].A meta-analysis by Boleij et al. [55] summarized the results of 52 case reports and 31 case series with a total of 189 patients with S. gallolyticus subsp gallolyticus and 151 patients with S. bovis biotype II bacteremia.The authors found that patients with S. gallolyticus subsp gallolyticus bacteremia had a higher risk of colorectal cancer compared with patients with S. bovis biotype II bacteremia.However, our meta-analysis included case-control and cohort studies, and we excluded case reports and case series.Overall, only 1 study [49] that was included in the meta-analysis by Boleij et al. [55] was included in our analysis as well, with the rest of the studies included adding new information and consolidating the association between SBSEC and colorectal neoplasia.Taken together, our meta-analysis included observational studies that assessed the risk of colorectal neoplasia in patients with evidence of SBSEC bacteremia, colonization, and antibody responses to SBSEC antigens.In contrast, the study by Boleij et al. [55] restricted their evaluation to the association between SBSEC clinical infection and colorectal cancer, which could explain the differential findings.
Interestingly, we found that IgG antibody responses to S. gallolyticus subsp gallolyticus antigens were more common in patients with colorectal cancer than controls.However, the  presence of such antibodies is not able to indicate acute, localized infection in the colon, and it is unknown whether antibody development for S. gallolyticus subsp gallolyticus antigens occurs only after bacterial entry into the bloodstream or even when the bacterium is still located in the colon [56].
Regardless of when antibody positivity occurs, documenting the time from antibody detection to colorectal cancer diagnosis may aid in a better understanding of the role of S. gallolyticus subsp gallolyticus in carcinogenesis.Colorectal cancer is the result of a sequence of events leading from normal tissue to adenoma development and then to carcinoma, which can last more than 10 years [57].
From the studies included in our analysis, Butt et al. prospectively assessed patients with and without IgG antibodies against S. gallolyticus subsp gallolyticus pili protein antigens for colorectal cancer development and found that serologic responses to S. gallolyticus subsp gallolyticus antigens were 1.36 times more common in patients who developed colorectal cancer, in whom the mean time from blood draw to cancer diagnosis was 3.4 years [21].The latter observation suggests that carcinogenesis had already begun in some patients with antibodies against S. gallolyticus subsp gallolyticus antigens.Similarly, a large prospective study of 4063 patients with colorectal cancer found that cancer risk was associated with IgG antibody detection against S. gallolyticus subsp gallolyticus pilus proteins within 10 years of blood draw, but not beyond 10 years of antibody positivity.Individuals diagnosed within 10 years of blood draw were also more likely to have precursor lesions in their colon [56].Taken together, these findings suggest that once IgG antibodies are detected in the serum, patients may already have colonic premalignant lesions.More prospective studies that delineate the natural history of these antibodies are required to establish whether seropositivity for S. gallolyticus subsp gallolyticus antigens can be used as a marker of underlying colorectal neoplasia.
Our results showed that patients with colorectal cancer, but not colorectal adenoma, were more likely to be colonized by SBSEC isolates compared with controls.Contrary to our findings, a previous meta-analysis assessing 515 patients with colorectal neoplasia (either adenoma or carcinoma) and 642 controls found that culture of SBSEC isolates from fecal specimens did not occur with greater frequency in patients with colorectal neoplasia compared with controls [26].Also, Boltin et al. prospectively assessed, over a period of 17 years, 15 individuals with and 100 individuals without evidence of SBSEC bacteria detected via culture in stool or colonic suction fluid and did not find a difference in colorectal cancer diagnosis between the groups [54].The difference in results between our analysis and the literature may be attributed to the fact that we included a sufficient number of patients colonized with SBSEC bacteria to detect a difference in colonization rates between patients with colorectal cancer and controls.Indeed, although only 1 [41] out of the 9 studies in our analysis found a significant association between SBSEC colonization and colorectal cancer, combining the results of the individual studies in our analysis found that patients with colorectal cancer were more likely to be colonized with SBSEC bacteria compared with controls.
Interestingly, our subgroup analysis found that colonization with S. gallolyticus subsp gallolyticus was not associated with underlying colorectal cancer diagnosis.Out of all members of the SBSEC, S. gallolyticus subsp gallolyticus has been described to adhere to malignant colorectal tissue with higher affinity than other SBSEC bacteria due to the unique property of its pili proteins binding exposed collagen fibers emerging from the basal lamina of malignant tissue [55,58,59].Moreover, Taddese et al. [60] found that S. gallolyticus subsp gallolyticus promotes tumorigenesis by modifying the ability of intestinal and cancerous cells to detoxify dietary components, thus allowing for DNA damage to accumulate, eventually increasing cancer risk [61,62].Kumar et al. also demonstrated that S. gallolyticus subsp gallolyticus enhanced cellular proliferation by upregulating beta-catenin, a pivotal molecule in colon tumorigenesis [63], while Oehmcke-Hecht et al. [64] found that S. gallolyticus subsp gallolyticus degraded tannic acid, which is a substance known to have cytotoxic effects against colorectal cancer cells [65].Failure to reveal a statistically significant association between S. gallolyticus subsp gallolyticus colonization and colorectal cancer can be explained by the small sample size used in our subgroup analysis or the involvement of other SBSEC bacteria in colorectal cancer development.Associations between SBSEC bacteria other than S. gallolyticus subsp gallolyticus and colorectal neoplasia stem from case reports and case series [11-13, 17, 51].Also, there are limited mechanistic studies available to provide plausible mechanisms by which SBSEC members other than S. gallolyticus subsp gallolyticus may preferentially colonize colorectal tissue [58].As a result, we highlight the importance of conducting studies with larger numbers of participants, in whom genotypic classification of SBSEC will occur, to establish the association between colonization with different SBSEC isolates and colorectal cancer.
Regarding the limitations of our analysis, only 2 studies reported underlying patient comorbidities [48,54] and only 3 studies had information about family history of colorectal cancer [22,23,54], so we could not assess whether these variables confound the relationship between the SBSEC and colorectal cancer.Genotypic classification of the SBSEC was not available in 9 studies included in the analysis [38,39,41,44,45,49,[52][53][54], and the remaining studies, with 1 exception [51], had information only about S. gallolyticus subsp gallolyticus, so comparison of the different subspecies with regards to risk of colorectal neoplasia was not possible.Next, colonization with SBSEC isolates was assessed via fecal culture evaluation in 7 studies [38,39,41,[43][44][45][46], and only 2 studies [36,42] examined colorectal tissue for SBSEC detection.As such, we can only conclude that colonization with SBSEC isolates was higher in patients with colorectal cancer compared with controls, but we cannot comment on whether SBSEC bacteria isolated from fecal or colonic tissue samples were more closely linked to colorectal cancer.Finally, serologic responses were examined against various S. gallolyticus subsp gallolyticus antigens, and we could not perform separate analyses for specific antigenic targets due to insufficient data.
In conclusion, our analysis reinforces results from previous studies that link SBSEC infection or colonization with enhanced colorectal cancer risk and associate serologic responses to S. gallolyticus subsp gallolyticus antigens with underlying colorectal neoplasia.Strengthening of these observations is necessary as SBSEC detection in the blood or isolation from fecal or colonic tissue can be used as a marker of underlying malignancy.Although evidence about the role of SBSEC in colorectal adenoma development is available in experimental and observational studies [48,49,53], it remains to be seen whether detection of premalignant lesions can be achieved through isolation of SBSEC bacteria in the blood or colorectal tissue.Well-designed prospective studies on the role of antibody responses to S. gallolyticus subsp gallolyticus antigens can also provide us with immunological markers for early detection of colorectal cancer.

Figure 1 .
Figure 1.Flow diagram for the selection of studies included in the systematic review and meta-analysis.

Figure 2 .
Figure 2. Individual and combined risk of colorectal neoplasia, either carcinoma (A) or adenoma (B), in patients with SBSEC bacteremia (with or without concomitant infective endocarditis) compared with individuals with no bacteremia diagnosis.The size of the squares is proportional to the weight of each study.Horizontal lines indicate the 95% CI of each study; diamond, the pooled estimate with 95% CI.Abbreviations: REML, restricted maximum likelihood; RR, relative risk; SBSEC, Streptococcus bovis/ Streptococcus equinus complex.

Figure 3 .
Figure 3. Individual and combined odds of colonization with SBSEC isolates in fecal or intestinal tissue in patients with colorectal neoplasia, either carcinoma (A) or adenoma (B), compared with individuals with no colorectal neoplasia diagnosis.The size of the squares is proportional to the weight of each study.Horizontal lines indicate the 95% CI of each study; diamond, the pooled estimate with 95% CI.Abbreviations: OR, odds ratio; REML, restricted maximum likelihood; SBSEC, Streptococcus bovis/ Streptococcus equinus complex.

Figure 4 .
Figure 4. Individual and combined odds of immunoglobulin G antibody responses to S. gallolyticus subsp gallolyticus antigens in patients with colorectal cancer compared with individuals with no colorectal cancer diagnosis.The size of the squares is proportional to the weight of each study.Horizontal lines indicate the 95% CI of each study; diamond, the pooled estimate with 95% CI.Abbreviations: OR, odds ratio; REML, restricted maximum likelihood; S. gallolyticus subsp gallolyticus, Streptococcus gallolyticus subsp gallolyticus.
used PCR in colorectal tissue for SBSEC detection.Out of these studies, 102 out of 401 (25.4%) patients with colorectal cancer and 48 out of 541 (8.9%) controls had evidence of SBSEC bacteria isolated from intestinal or fecal tissue samples.Patients with colorectal cancer were 2.27 times more likely to have evidence of colonization with SBSEC isolates compared with individuals with no colorectal cancer diagnosis (OR, 2.27; 95% CI, 1.11-4.62;I 2 = 46.54%)(Figure

Table 1 . Continued
Abbreviations: IgG, immunoglobulin G; NA, not applicable/available; PCR, polymerase chain reaction; SBSEC, Streptococcus bovis/Streptococcus equinus complex; S. gallolyticus subsp gallolyticus, Streptococcus gallolyticus subsp gallolyticus.a Sites for SBSEC detection included fecal and/or colorectal tissue.b When no phenotypic or genotypic classification was available, the general term SBSEC was used.