The effect of fecal bile acids on the incidence and risk-stratification of colorectal cancer: an updated systematic review and meta-analysis

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-84801-6.

Keywords: Colorectal cancer, Adenoma, Bile acid, Meta-analysis

Subject terms: Colorectal cancer, Tumour biomarkers

Literature search

An online search identified relevant cross-sectional and case-control studies published online in the major English language databases (Medline, Embase, Web of Science, AMED, and CINAHL). The papers search for this study included only English peer-reviewed articles published before January 1st, 2024.

Steps for searching: (I) Search for articles, systematic reviews, and meta-analyses from the databases, and (II) Analyze and summarize subject words and keywords. The search terms were: “Bile Acids,” “Bile Salts,” and “Colorectal Neoplasms,” “Colonic Neoplasms,” “colorectal cancer” or “colon cancer,” “colorectal carcinoma” or “colon carcinoma,” “colorectal neoplasm,” “colon neoplasm,” “colorectal neoplasia” or “colon neoplasia,” “Colorectal Carcinomas.” Details of the search strategy: ((Bile Acid) OR (Bile Salt)) AND ((Colorectal Neoplasms) OR (Colonic Neoplasms) OR (colorectal cancer) OR (colon cancer) OR (colorectal carcinoma) OR (colon carcinoma) OR (colorectal neoplasm) OR (colon neoplasm) OR (colorectal neoplasia) OR (colon neoplasia) OR (Colorectal Carcinomas)) NOT ( (rats) OR (mouse)) OR (mice) OR (murine)) AND (humans [Filter]) (III) Further search for references and associated papers. (IV) Screening of articles according to the inclusion and exclusion criteria.

Inclusion and exclusion criteria

Cross-sectional and case-control studies of fecal BA levels in patients with CRC or those at high risk for CRC have been officially published as of 1 January 2024. The studies used similar objectives and methods to compare fecal BA levels in patients with CRC or those at high risk for CRC with healthy people. The case-control study consisted of patients with CRC confirmed by endoscopic pathology as a case group and patients without CRC as a control group. Information on their fecal BA content was collected and analyzed. In the cross-sectional study, patients at high risk for CRC identified through epidemiological investigation served as the case group.

In contrast, a comparable group of non-high-risk CRC patients within the same time and scope was risk colorectal cancer patients within the same time and scope were selected as the control group. The primary outcome of the studies was to investigate the contents and types of various BAs in the feces of the case or control groups. The local ethics committee approved all studies, and informed consent was obtained.

All studies were required to have original paper and extractable data available. For duplicate publications, smaller datasets (the total number of subjects is less than ten), studies with incomplete or contradictory information, and significant errors in statistical methods were excluded. Reviews, letters, case reports, animal studies, and studies with non-primary data were also excluded.

Literature screening and data extraction

We imported all search records (including author names, dates, journal titles, and abstracts) into EndNote X7 and deleted duplicate records. Search results were screened according to “inclusion criteria” and “exclusion criteria,” and then basic experiments and animal experiments were excluded. Two researchers (Shaohui Yang and Yu Wang) did the study screening independently, and a third researcher (Wei Cui) would decide if there was disagreement. Finally, researchers (Lijuan Sheng and Chenyang Ma) read the study. They extracted the primary study information (e.g., author, year, region), study object information (e.g., number, age, gender), and detection indicators (e.g., detection method, fecal BA type, content).

Literature quality evaluation

Two researchers evaluated study quality independently, and a third researcher would make the final decision if there were disagreements. The Newcastle-Ottawa Scale (NOS) was used for case-control and cross-sectional studies to evaluate study quality¹². The difference was that cross-sectional studies used modified NOS for evaluation¹³. NOS scores range from 0 to 9; studies with 0–3, 4–6, and 7–9 were considered low, moderate, and high-quality, respectively. Data from studies with a high risk of bias were unreliable, so they were excluded.

Statistical methods

The meta-analysis was performed using the RevMan 5.3 statistical software provided by www.cochrane.org. Various BA components were analyzed by subgroup analysis, and then the results of each subgroup analysis were combined for analysis. Because of the possible heterogeneity, a random effects model was applied. A fixed-effect model was used to analyze data with no significant heterogeneity (I² < 50%). The data in each study were recorded differently and converted to “mean ± standard deviation” form before the analysis began. The expression of concentrations varied across studies; therefore, SMD was used for analysis. A funnel plot was drawn to detect publication bias. When p < 0.05, the difference was statistically significant.

One article¹⁴ had no calculable data (mean, standard deviation, standard error, or 95% CI) and only provided a statistical graph. We use online analytical tools (https://automeris.io/WebPlotDigitizer/) to extract the data from the statistical figures. Some articles had no specific data or statistical graphs, so they are considered qualitative analyses.

Literature search results

Quality evaluation of the study

Different NOS were used to evaluate the quality of 19 case-control^15–33 and four cross-sectional^14,34–36 studies. According to the scoring criteria, the quality evaluation results of 19 case-control studies^15–33 were moderate and high. The quality evaluation results of four cross-sectional studies [14, 34–36] were all high-quality, showing that their data results are reliable (Tables 1 and 2).

Characteristics of the selected study

The studies were divided into quantitative and qualitative studies according to the data availability for calculation (Table 3). Among the quantitative studies, three were cross-sectional studies^14,34,36 (Table 3), and the remainder were case-control studies. One study was a cross-sectional study³⁵ (Table 3), and five were case-control studies^{15,19,20,28,31}, among six qualitative studies^{15,19,20,28,31,35}. Cross-sectional studies compared people at high and low risk of CRC, so we identified all cross-sectional studies as CRC risk categories. In the case-control studies, seven^{20,22,27–30,33} were CRC incidence category due to the comparison of CRC cases and healthy controls, 11 studies^{15,17–19,21,23–26,31,32} were considered a CRC risk category because of “adenomatous polyps patients vs. healthy controls,” and one study comparing CRC cases and adenomatous polyps patients with healthy controls were considered as CRC incidence/risk categories¹⁶. These studies were conducted in multiple regions and countries (including China, the United Kingdom, the United States, France, Japan, Germany, and Zimbabwe) and included 1,265 patients and healthy people. All subjects had explicit inclusion and exclusion criteria. Endoscopy and pathology identified and diagnosed CRC/adenomatous polyp cases and healthy controls. In all case-control studies, the BA conditions of the case and control groups were described to ensure they were comparable. In cross-sectional studies, two similar populations were compared at a specific time and within a particular area to find the cause.

Due to different eras, all studies used different detection methods and technologies, including gas chromatography, mass spectrometry, and gas-liquid chromatography. Therefore, the accuracy of various detection techniques was different, but the meta-analysis results were still valuable. The chemical components tested in each study were cholic acid, chenodeoxycholic acid, deoxycholic acid, lithocholic acid, and ursodeoxycholic acid. Primary bile, secondary bile, and total BAs were also included, but they were analyzed separately (Table 3).

Data analyses

The meta-analysis of the quantitative data extracted from the studies was presented in Figs. 2A–C and 3A–C. In the CRC risk meta-analysis, the effect size of CA, CDCA, DCA, and UDCA were significantly higher (CA: SMD = 0.41, 95% CI: 0.5–0.76, P = 0.02; CDCA: SMD = 0.35, 95% CI: 0.09–0.62, P = 0.009; DCA: SMD = 0.33, 95% CI: 0.03–0.64, P = 0.03; UDCA: SMD = 0.46,95% CI: 0.14–0.78, P = 0.005), and the combined effect size was significantly higher in high-risk than the low-risk CRC group (SMD = 0.36, 95% CI: 0.21–0.51, P < 0.00001) (Fig. 2A). The effect size of primary BAs was significantly higher (SMD = 0.44, 95% CI: 0.09–0.71 P = 0.01); however, the combined effect size of primary and secondary BAs was not significantly higher (Fig. 2B). The effect size of total BAs was not significantly higher in the high-risk vs. the low-risk group (Fig. 2C). Total BA means collecting all the BA molecules, not only CA, CDCA, DCA, LCA, and UDCA, as were primary and secondary BAs.

In CRC incidence meta-analysis, the effect size of CA and CDCA were significantly higher (CA: SMD = 0.42, 95% CI: 0.04–0.80, P = 0.03; CDCA: SMD = 0.61,95% CI: 0.26–0.96, P = 0.00079), and their combined effect size was also significantly higher in high-risk compared to low-risk CRC group (SMD = 0.39, 95% CI: 0.09–0.68, P = 0.01) (Fig. 3A). The effect size of primary, secondary, and total BAs (including the combined effect size of primary and secondary BAs) all were not significantly higher in the high- vs. low-risk group (Fig. 3B, C).

Table 4 presents all the results of the random-effect model meta-analyses. Data with low heterogeneity were more suitable for the fixed-effect model³⁷; therefore, we performed another meta-analysis using a fixed-effect model for data with I² less than 50%, and the results showed the same conclusion as the random-effect model (Table 5).

In qualitative studies, the results of only one cross-sectional study³⁵ suggested that the concentration of CDCA, DCA, and UDCA in the stool of the CRC high-risk group than in the low-risk group, similar to Hill et al.¹⁵. However, Weir et al. suggested that the level of UDCA in the stool of HC was higher. In addition, the results of the other three cross-sectional and case-control studies did not show differences in fecal BA levels between CRC high-risk vs. low-risk groups or CRC vs. HC. Some of the qualitative analysis results were consistent with our meta-analysis.

Publication bias detection

We used a funnel plot to detect publication bias for several primary BAs in stool, and the results are shown in Fig. 4. The asymmetric distribution of each point in the funnel plot suggested the presence of publication bias in this study.

Competing interests

The authors declare no competing interests.

• 1.Sung, H. et al. Global Cancer statistics 2020: GLOBOCAN estimates of incidence and Mortality Worldwide for 36 cancers in 185 countries. CA Cancer J. Clin.71, 209–249 (2021). [DOI] [PubMed] [Google Scholar]
• 2.Click, B., Pinsky, P. F., Hickey, T., Doroudi, M. & Schoen, R. E. Association of Colonoscopy Adenoma findings with Long-Term Colorectal Cancer incidence. JAMA319, 2021–2031 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.Strum, W. B. & Colorectal Adenomas N Engl. J. Med.374, 1065–1075 (2016). [DOI] [PubMed] [Google Scholar]
• 4.Chen, H. et al. Urea cycle activation triggered by host-microbiota maladaptation driving colorectal tumorigenesis. Cell Metabol.35, 651–666e657 (2023). [DOI] [PubMed] [Google Scholar]
• 5.Tong, J. L., Ran, Z. H., Shen, J., Fan, G. Q. & Xiao, S. D. Association between fecal bile acids and colorectal cancer: a meta-analysis of observational studies. Yonsei Med. J.49, 792–803 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 6.Wang, S. et al. Interplay between bile acids and the gut microbiota promotes intestinal carcinogenesis. Mol. Carcinog.58, 1155–1167 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 7.Collins, S. L., Stine, J. G., Bisanz, J. E., Okafor, C. D. & Patterson, A. D. Bile acids and the gut microbiota: metabolic interactions and impacts on disease. Nat. Rev. Microbiol.21, 236–247 (2023). [DOI] [PubMed] [Google Scholar]
• 8.Fuchs, C. D. & Trauner, M. Role of bile acids and their receptors in gastrointestinal and hepatic pathophysiology. Nat. Rev. Gastroenterol. Hepatol.19, 432–450 (2022). [DOI] [PubMed] [Google Scholar]
• 9.Yachida, S. et al. Metagenomic and metabolomic analyses reveal distinct stage-specific phenotypes of the gut microbiota in colorectal cancer. Nat. Med.25, 968–976 (2019). [DOI] [PubMed] [Google Scholar]
• 10.Devkota, S. et al. Dietary-fat-induced taurocholic acid promotes pathobiont expansion and colitis in Il10-/- mice. Nature487, 104–108 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 11.Wu, Y., Mo, R., Zhang, M., Zhou, W. & Li, D. Grape seed Proanthocyanidin alleviates intestinal inflammation through gut microbiota-bile acid crosstalk in mice. Front. Nutr.8, 786682 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 12.Alvandi, E., Wong, W. K. M., Joglekar, M. V., Spring, K. J. & Hardikar, A. A. Short-chain fatty acid concentrations in the incidence and risk-stratification of colorectal cancer: a systematic review and meta-analysis. BMC Med.20, 323 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 13.Modesti, P. A. et al. Panethnic differences in blood pressure in Europe: a systematic review and Meta-analysis. PLoS One. 11, e0147601 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 14.Ou, J. et al. Diet, microbiota, and microbial metabolites in colon cancer risk in rural africans and African americans. Am. J. Clin. Nutr.98, 111–120 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 15.Hill, M. J. et al. Faecal bile-acids and clostridia in patients with cancer of the large bowel. Lancet1, 535–539 (1975). [DOI] [PubMed] [Google Scholar]
• 16.Reddy, B. S. & Wynder, E. L. Metabolic epidemiology of colon cancer. Fecal bile acids and neutral sterols in colon cancer patients and patients with adenomatous polyps. Cancer39, 2533–2539 (1977). [DOI] [PubMed] [Google Scholar]
• 17.Murray, W. R., Backwood, A., Trotter, J. M., Calman, K. C. & MacKay, C. Faecal bile acids and clostridia in the aetiology of colorectal cancer. Br. J. Cancer. 41, 923–928 (1980). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 18.Kaibara, N., Sasaki, T., Ikeguchi, M., Koga, S. & Ikawa, S. Fecal bile acids and neutral sterols in Japanese with large bowel carcinoma. Oncology-basel40, 255–258 (1983). [DOI] [PubMed] [Google Scholar]
• 19.Hikasa, Y., Tanida, N., Ohno, T. & Shimoyama, T. Faecal bile acid profiles in patients with large bowel cancer in Japan. Gut25, 833–838 (1984). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 20.Tanida, N., Hikasa, Y., Shimoyama, T. & Setchell, K. D. Comparison of faecal bile acid profiles between patients with adenomatous polyps of the large bowel and healthy subjects in Japan. Gut25, 824–832 (1984). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 21.Breuer, N. F., Dommes, P., Jaekel, S. & Goebell, H. Fecal bile acid excretion pattern in colonic cancer patients. Dig. Dis. Sci.30, 852–859 (1985). [DOI] [PubMed] [Google Scholar]
• 22.Breuer, N. F., Jaekel, S., Dommes, P. & Goebell, H. Fecal bile acids in patients with adenomatous polyps of the colon. Case-control study. Digestion34, 87–92 (1986). [DOI] [PubMed] [Google Scholar]
• 23.Owen, R. W., Henly, P. J., Thompson, M. H. & Hill, M. J. Steroids and cancer: faecal bile acid screening for early detection of cancer risk. J. Steroid Biochem.24, 391–394 (1986). [DOI] [PubMed] [Google Scholar]
• 24.Hill, M. J., Melville, D. M., Lennard-Jones, J. E., Neale, K. & Ritchie, J. K. Faecal bile acids, dysplasia, and carcinoma in ulcerative colitis. Lancet2, 185–186 (1987). [DOI] [PubMed] [Google Scholar]
• 25.Owen, R. W., Dodo, M., Thompson, M. H. & Hill, M. J. Fecal steroids and colorectal cancer. Nutr. Cancer. 9, 73–80 (1987). [DOI] [PubMed] [Google Scholar]
• 26.Korpela, J. T., Adlercreutz, H. & Turunen, M. J. Fecal free and conjugated bile acids and neutral sterols in vegetarians, omnivores, and patients with colorectal cancer. Scand. J. Gastroenterol.23, 277–283 (1988). [DOI] [PubMed] [Google Scholar]
• 27.Owen, R. W., Day, D. W. & Thompson, M. H. Faecal steroids and colorectal cancer: steroid profiles in subjects with adenomatous polyps of the large bowel. Eur. J. Cancer Prev.1, 105–112 (1992). [PubMed] [Google Scholar]
• 28.de Kok, T. M. et al. Bile acid concentrations, cytotoxicity, and pH of fecal water from patients with colorectal adenomas. Dig. Dis. Sci.44, 2218–2225 (1999). [DOI] [PubMed] [Google Scholar]
• 29.Meance, S. et al. Fecal primary bile acids and serum cholesterol are associated with colorectal adenomas. Dig. Dis. Sci.48, 1751–1757 (2003). [DOI] [PubMed] [Google Scholar]
• 30.Boutron-Ruault, M. C. et al. Effects of a 3-mo consumption of short-chain fructo-oligosaccharides on parameters of colorectal carcinogenesis in patients with or without small or large colorectal adenomas. Nutr. Cancer. 53, 160–168 (2005). [DOI] [PubMed] [Google Scholar]
• 31.Weir, T. L. et al. Stool microbiome and metabolome differences between colorectal cancer patients and healthy adults. PLoS One. 8, e70803 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 32.Torii, T., Kanemitsu, K. & Hagiwara, A. Simultaneous assay of fecal short-chain fatty and bile acids and ratio of total bile acids to butyrate in Colon cancer. CHROMATOGRAPHY40, 49–57 (2019). [Google Scholar]
• 33.Chen, C. et al. Bacteroides, butyric acid and t10,c12-CLA changes in colorectal adenomatous polyp patients. Gut Pathog. 13, 1 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 34.Ou, J., DeLany, J. P., Zhang, M., Sharma, S. & O’Keefe, S. J. Association between low colonic short-chain fatty acids and high bile acids in high colon cancer risk populations. Nutr. Cancer. 64, 34–40 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 35.Katsidzira, L. et al. Differences in fecal gut microbiota, short-chain fatty acids and bile acids Link Colorectal Cancer Risk to Dietary Changes Associated with Urbanization among zimbabweans. Nutr. Cancer. 71, 1313–1324 (2019). [DOI] [PubMed] [Google Scholar]
• 36.Ocvirk, S. et al. A prospective cohort analysis of gut microbial co-metabolism in Alaska native and rural African people at high and low risk of colorectal cancer. Am. J. Clin. Nutr.111, 406–419 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 37.Tan, E., Braithwaite, I., McKinlay, C. J. D. & Dalziel, S. R. Comparison of Acetaminophen (paracetamol) with Ibuprofen for Treatment of Fever or Pain in children younger than 2 years: a systematic review and Meta-analysis. JAMA Netw. open.3, e2022398 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 38.Cai, J., Sun, L. & Gonzalez, F. J. Gut microbiota-derived bile acids in intestinal immunity, inflammation, and tumorigenesis. Cell. host Microbe. 30, 289–300 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 39.Capron, J. P., Delamarre, J., Canarelli, J. P., Brousse, N. & Dupas, J. L. Does cholecystectomy predispose to colo-rectal cancer? Gastroenterol. Clin. Biol.2, 383–389 (1978). [PubMed] [Google Scholar]
• 40.Shao, T. & Yang, Y. X. Cholecystectomy Risk Colorectal Cancer100, 1813–1820 (2005). [DOI] [PubMed] [Google Scholar]
• 41.Ren, X. et al. Bacterial alterations in Post-cholecystectomy patients are Associated with Colorectal Cancer. Front. Oncol.10, 1418 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 42.Sarashina-Kida, H. et al. Gallbladder-derived surfactant protein D regulates gut commensal bacteria for maintaining intestinal homeostasis. Proc. Natl. Acad. Sci. U S A. 114, 10178–10183 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 43.Zwicker, B. L. & Agellon, L. B. Transport and biological activities of bile acids. Int. J. Biochem. Cell. Biol.45, 1389–1398 (2013). [DOI] [PubMed] [Google Scholar]
• 44.Sheng, W., Ji, G. & Zhang, L. The Effect of Lithocholic Acid on the gut-liver Axis. Front. Pharmacol.13, 910493 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 45.Jia, W., Xie, G. & Jia, W. Bile acid-microbiota crosstalk in gastrointestinal inflammation and carcinogenesis. Nat. Rev. Gastroenterol. Hepatol.15, 111–128 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 46.Wirbel, J. et al. Meta-analysis of fecal metagenomes reveals global microbial signatures that are specific for colorectal cancer. Nat. Med.25, 679–689 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 47.Kühn, T. et al. Prediagnostic plasma bile acid levels and Colon cancer risk: a prospective study. J. Natl. Cancer Inst.112, 516–524 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 48.Sun, L. et al. Bile salt hydrolase in non-enterotoxigenic Bacteroides potentiates colorectal cancer. Nat. Commun.14, 755 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 49.Li, T. et al. A gut microbiota-bile acid axis promotes intestinal homeostasis upon aspirin-mediated damage. Cell. host Microbe. 32, 191–208e199 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 50.Jiang, Z. et al. The gut microbiota-bile acid axis links the positive association between chronic insomnia and cardiometabolic diseases. Nat. Commun.13, 3002 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 51.Honda, A. et al. Regulation of bile acid metabolism in mouse models with hydrophobic bile acid composition. J. Lipid Res.61, 54–69 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 52.Pan, Y. et al. Novel approaches in IBD therapy: targeting the gut microbiota-bile acid axis. Gut Microbes. 16, 2356284 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 53.Kastl, A. et al. Dietary fiber-based regulation of bile salt hydrolase activity in the gut microbiota and its relevance to human disease. Gut Microbes. 14, 2083417 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 54.Sannasiddappa, T. H., Lund, P. A. & Clarke, S. R. Vitro Antibacterial Activity of Unconjugated and conjugated bile salts on Staphylococcus aureus. Front. Microbiol.8, 1581 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 55.do Nascimento, P. G. et al. Lithocholic acid and derivatives: antibacterial activity. Steroids104, 8–15 (2015). [DOI] [PubMed] [Google Scholar]
• 56.Qi, X. et al. Gut microbiota-bile acid-interleukin-22 axis orchestrates polycystic ovary syndrome. Nat. Med.25, 1225–1233 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 57.Cheung, Y. B., Ma, X., Lam, K. F., Li, J. & Milligan, P. Bias control in the analysis of case-control studies with incidence density sampling. Int. J. Epidemiol.48, 1981–1991 (2019). [DOI] [PubMed] [Google Scholar]

Supplementary Materials

Data Availability Statement

Some or all data are available from the corresponding author by request.