The Relationship Between Gut Microbiome Estrobolome and Breast Cancer: A Systematic Review of Current Evidences

Abstract

Breast cancer is the most frequent kind of cancer and the second leading cause of mortality worldwide, behind heart disease. Next-generation sequencing technologies enables for unprecedented enumeration of human resident gut microorganisms, conferring novel insights into the role of the microbiota in health and individuals with breast cancer. A growing body of research on microbial dysbiosis seems to indicate an elevated risk of health complications including cancer. Although several dysbiosis indices have been proposed, their underlying methodology, as well as the cohorts and conditions of breast cancer patients are significantly different. To date, these indices have not yet been thoroughly reviewed especially when it comes to researching the estrogen-gut microbiota axis. Instead of providing a thorough rating of the most effective diversity measurements, the current work aims to be used to assess the relevance of each study's findings across the demographic data, different subtypes, and stages of breast cancer, and tie them to the estrobolome, which controls the amount of oestrogen that circulates through humans. This review will cover 11 studies which will go into a detailed discussion for the microbiome results of the mentioned studies, leaving to the user the final choice of the most suited indices as well as highlight the observed bacteria found to be related to the estrobolome in hopes of giving the reader a better understanding for the biological cross-talk between gut microbiome and breast cancer progression.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12088-023-01135-z.

Introduction

Breast cancer is the leading cause of global cancer incidence in 2022, with an estimate of 287,850 cases were diagnosed in women [59]. It is the fifth leading cause of mortality worldwide estimated around 685,000 deaths [56]. While the incidence is higher (> 80 per 100,000) in developed countries such as Australia, Western Europe, North America, and North Europe, the rate is lower within developing countries (< 40 per 100,000) such as Central America, Middle Africa, and Southeast Asia [1, 56]. Human breast cancer is a highly heterogeneous illness, and there is a lot of variation in how it develops, progresses, and metastasizes even within particular subtypes of the disease [58].

Although a mix of genetic, epigenetic, and environmental variables has been identified, the actual cause of breast cancer is still unclear despite extensive study was performed. Inherited deletion of the BRCA1/2 gene explains just a small portion of breast cancer cases, however the majority of sporadic breast cancer (up to 70%) is still unknown [19]. Lifestyle factors such as diet [29, 42], alcohol, and radiation may play a role in the sporadic etiology [28, 29]. Endogenous estrogen levels and estrogen metabolism changes have been associated to an increased risk of breast cancer, especially among postmenopausal women [50]. This raises the question of whether the pathogenesis of breast cancer is shaped by specific microbes such as gut bacteria, viruses or fungi. The gut microbiome is the group of bacteria that live together as a community in the host’s gut. Dysbiosis, on the other hand, refers to a condition of imbalance within the microbiome [2]. Gut microbiota dysbiosis is an emerging field of research that seems to be linked to an increased risk of developing inflammatory, autoimmune, and malignant diseases [3] including breast cancer [4]. In 15–20% of cases, microorganisms have been linked to the development of cancer, according to studies on risk factors for the disease. Numerous connections between illness and the gut microbiota are linked to the particular microbe types involved in the development of disease as well as the microbiota's makeup.

Alterations in the typical balance of gut microbiota are linked to an increased risk of developing breast cancer. These changes could potentially impact the body's internal estrogen levels or disrupt estrogen processing, ultimately contributing to the development of breast cancer in the absence of a genetic predisposition. To date, obesity and a high-fat diet, especially in cases of sporadic breast cancer may also be linked to the microbial communities within the host as a factor breast carcinogenesis [8, 21, 58]. The development, treatment, and prognosis of breast cancer patients, may all be impacted by gut dysbiosis, either directly or indirectly. The notion of the estrobolome, which is a group of bacteria responsible for estrogen regulation within their hosts, was recently introduced. Studies have shown that certain types of gut bacteria can affect estrogen metabolism and may influence the risk of breast cancer. For example, some gut bacteria can produce an enzyme called β-glucuronidase (GUS) [55], which can increase the levels of circulating estrogen. In addition, some gut bacteria can produce compounds that have estrogenic effects, which can also contribute to breast cancer development. Notable patterns of gut microbiota in breast cancer patients revealed abundance of Firmicutes and Bacteroidetes phyla [63, 66] that predominantly classified as gut microbial β-glucuronidase (gmGUS) [55, 61]. Despite the fact that research into the precise connections between the gmGUS, GUS, and estrogen metabolism is still ongoing, several studies have demonstrated the link. Briefly, GUS is an enzyme that breaks down glycosidic bonds, releasing aglycones from glycosides. In the gastrointestinal tract, glucuronide metabolites formed by the liver can be deconjugated by gmGUS in the GI tract. This process is important for the growth of gut microbiota and chemical biotransformation. Estrogen, an endogenous aglycone, is metabolized into glucuronide in the liver and deconjugated by gmGUS in the gut. The reactivation of estrogen by gmGUS is hypothesized to play a role in microbiota-host interactions and may be linked to estrogen-related diseases like breast cancer.

This evidence led to the hypothesis that gut dysbiosis may have an impact on the estrobolome and could contribute to the development of breast cancer [19]. Although there have been some prior efforts to link gut microbiota dysbiosis with breast cancer development [43, 47], there is no established standard for assessing dysbiosis [51], as the gut microbiome may differ amongst populations due to variations in environmental factors, diet, social background, and lifestyle [52]. Therefore, we have conducted an extensive assessment by evaluating the utility of alpha- and beta-diversities from both older and the most recent studies up to the year 2023. The diversity and specific species that make up the human gut microbiota are very complicated, remarkably individualistic, and exists in a dynamic equilibrium [28]. Once microbial abundances were determined using next-generation sequencing (NGS), diversity indices offered valuable mathematical approaches to describe the ecological complexity of a single sample or to discover species differences between samples [20]. The terms “α-diversity” refers to the average species variety in sites within a specific group, and “β-diversity” relates to the differentiation between different groups [57]. The conventional approach is to employ separate α- and β-diversity indices to assess intra- and inter-site diversity [32]. The major lesson is that these diversity matrices assist to emphasise the importance of the gut microbiome identified in breast cancer patients and helps relate those findings to understanding the progression of breast cancer.

The objectives of this systematic review are to delineate patterns in the gut microbiota profiles of breast cancer patients and healthy populations, and to determine their potential use as biomarkers for the early detection and diagnosis of illness. This is accomplished by comparing significant amounts of bacteria from breast cancer patients to known sources of the enzymes GUSs [41, 55]. In addition, this systematic review aimed to provide a brief summary of the major indices for analysing the diversity pattern of the human gut microbiota found within breast cancer patients and to related its importance to demographic data, different subtypes, and stages of breast cancer. The findings of this review may pave the way for additional research that may suggest a role for gut microbiome manipulation in the management of breast cancer or as potential tumor marker to predict the course or prognosis of the disease. A rapid, non-invasive diagnosis strategy or enhanced cancer management techniques will be suggested in the near future.

Materials and Methods

Data Sources and Search Strategy

The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) 2020 guidelines were used to guide on the reporting of this systematic review [40]. We retrieved primary articles from PubMed, Web of Science, SCOPUS and Science Direct. The articles were retrieved from their inception to October 2023. A string of search terms; intestinal microbiota, intestinal microbiome, gut microbiome, gut microbiota, gastrointestinal microbiota, gastrointestinal microbiome, faecal microbiota, faecal microbiome, gut microbiome dysbiosis, intestinal microbiome dysbiosis, gut microbiota dysbiosis, gut microbiome metabolome, gut microbial metabolome, gut metabolome, faecal metabolome, estrobolome, β-glucuronidase, gut microbiota profiling, gut microbiome profiling, faecal microbiome profiling, gut microbial metabolism, bacterial metabolite and breast cancer were screened. All search results were filtered by language (English only). The bibliographic records from each database were exported to EndNote X9 (Thomson Reuters, New York, NY) where duplicates were removed.

Study Selection and Data Extraction

All references were first categorized using a Smart Group function in EndNoteX9 that filtered all abstract contains gut, microbe and breast cancer. Three reviewers, ASMY, NSK and NEM independently screened titles and abstracts for eligibility based on the selection criteria before evaluating full texts. Any disagreements or discrepancies were discussed with NSK, HFA, NSA, DDT, NMM and NEM and/or resolved by consensus. Reports were excluded from this review if they were review, abstract without full text available, or unpublished referenced reports, as well as non-English articles were excluded.

Quality Assessment

The Mixed Methods Assessment Tool (MMAT) [26] was used to appraise the quality of included study. The overall quality assessment of the studies was evaluated independently by NEM, DDT, ASMY, and HFA. For each aspect of the quality assessment, the overall results of the appraisal were interpreted as the lowest score of the study components met the criteria. All study results were presented in tables and figures using narrative format.

Results

Literature Search for Gut Microbiome in Breast Cancer

In total, 897 reports were identified by the search of databases. After removing duplicates (n = 715) and abstract not associated with the gut, microbiome, and breast cancer (n = 407) were excluded. 308 studies were screened based on titles and abstracts. 203 full text articles were reviewed for eligibility. Finally, 11 studies were included in this systematic review (Fig. 1).

Fig. 1.

PRISMA flowchart of literature search and inclusion

Study Characteristic

Detail information of the studies that met inclusion criteria are presented in Table 1. Compulsory characteristics of included studies for this review include; (1) study of gut microbiomes in breast cancer patients compared to healthy control, (2) feces samples were collected before the patients receive any treatments, (3) specimen technique must be from feces only, (4) a method used for gut microbiome characterization and quantification must be at least Illumina Next-Generation Sequencing (NGS) targeting on V3 or V4 or V3 and V4 region of 16s rRNA, (5) focus on reporting significance contribution of gut microbiome in breast cancer case through alpha- and beta-diversity metrics assessment, and (6) reported significant bacteria coincide with known producers of GUSs. Studies that involve other factors that may attribute to the change in gut microbiome diversity in breast cancer patients, for example; change of diet, consumption of medication, participation in vigorous physical activities among breast cancer patients were excluded from this review. Studies of gut microbiome diversity of which feces specimen only collected from healthy individuals or breast cancer cases were also excluded from this report.

Table 1.

Characteristics of reviewed studies

References	Country	Sample size		Mean age (years)		Purpose	Study design	Specimen technique	Method	Alpha diversity metrics	Beta diversity metrics
		Breast cancer cases	Controls	Breast cancer cases	Controls
Jiang et al. [30]	Hainan Province, China	20 Menopausal 21 Premenopausal 2 Hysterectomy	23 Menopausal 7 Premenopausal	52.9	61.6	Analyzed the intestinal microbial composition and functional metabolism of breast cancer patients in Hainan Province, China	Case–control study	Feces	PacBio sequencing platform targeting 16S rRNA gene using amplification primer 27F and 1492R	Richness: Chao1 Ace Diversity: Shannon index	Jaccard distance
Ma et al. [38]	Xining, China	26 Breast Cancer patients 20 Benign breast lesion patients	20	Not specifed	Not specified	Studied the gut microbiota of patients with breast cancer and benign breast lesions using 16S rRNA gene sequencing	Case–control study	Feces	Illumina HiSeq 2500 targeting V3-V4 region	Observed species index, Chao1	Unweighted Unifrac, and weighted Unifrac
Wu et al. [64]	Ghana, Africa	369 Breast Cancer patients 93 Non-malignant cases	419	Not specified Range: 18–74	Not specified Range: 18–74	Analyzed the link between the oral microbiome and breast cancer as well as non-malignant breast disease within Ghana population using 16S rRNA gene sequencing	Case–control study	Feces and saliva (oral)	Illumina MiSeq sequencing targeting V4 region	Richness: Observed ASV Diversity: Faiths’s phylogenetic diversity, and Shannon index	Bray–Curtis, Jaccard, unweighted UniFrac, and weighted UniFrac distances
Yao et al. [65]	Guangdong, China	19 Sleep distrubance patients 17 No sleep distrubance patients	No healthy cohort reported	46.95 Sleep disturbance 49.18 No sleep distubance	No healthy cohort reported	Studied the potential relationship between gut microbiota, sleep disturbances, and acute postoperative pain in breast cancer patients	Case–control study	Feces	Illumina HiSeq sequencing targeting V3-V4 region	Richness: Observed ASV Diversity: Shannon index Simpson index	Weighted Unifrac
Byrd et al. [9]	Ghana, Africa	379 Breast cancer cases 102 Non-malignant breast disease cases (Before any treatment)	414	Breast cancer cases = 54.10 Non malignant cases = 38.80	46.90	Explored associations of the fecal microbiota with breast cancer and non-malignant breast disease in Sub-Saharan Africa	Case–control study	Feces	Illumina MiSeq targeting on the V4 region of the 16S rRNA gene	Richness: Observed (ASV) Diversity: Shannon diversity index, and Faith’s phylogenetic diversity	Bray Curtis, Unweighted/weighted UniFrac distance matrices
He et al. [25]	Southern Anhui, China	54 Premenopausal breast cancer cases	28	39.74	37.54	Identified the intestinal flora of breast cancer patients in Southern Anhui using targeted metabolomic studies and 16 s rRNA gene sequencing	Case–control study	Feces and breast cancer tissue	Illumina sequencies targeting the V3-V4 region of the 16S rRNA gene	Richness: Pielou’s evenness Diversity: Shannon index, Simpson index	Redunancy Analysis (RDA)
Hou et al. [27]	Kaohsiung, Taiwan	100 Postmenopausal 100 Premenopausal	67 Controls (50-premenopausal, 17-postmenopausal)	41.5 Premenopausal 60.08 Postmenopausal	35.4 Premenopausal 61.6 Postmenopausal	Investigated the gut microbiota profiles of premenopausal breast cancer patients to further clarify related microbial markers, diagnostic values as well as functional pathway	Case–control study	Feces	Illumina MiSeq sequencing targeting the V3-V4 region of the 16S rRNA gene	Shannon index	Unifrac distance matrice
Bobin-Dubigeon et al. [7]	Saint-Herblain, France	25 (before any anticancer therapy was started)	30	63.0	53.5	Investigated the gut microbiota of female breast cancer patient’s vs healthy individuals in France population as a preliminary study	Case–control study	Feces	Illumina MiSeq sequencing targeting the V3-V4 region of the 16S rRNA gene	Richness: Rarefied Chao1 Diversity: Shannon index	Unidentified Principle Coordinates Analysis
Wu et al. [63]	USA	20 Premenopausal 17 Postmenopausal	No healthy cohort reported	50.6	No healthy cohort reported	Explored the association between breast tumor characteristics and the gut microbiome as well as known breast cancer risk factors, in incident breast cancer patients	Case Study	Feces	Illumina MiSeq sequenicng targeting V3 and V4 region	Richness: Observed species index, Chao 1, and phylogenetic distance Diversity: Shannon index	Weighted Unifrac and Unweighted Unifrac
Zhu et al. [67]	Guangxi, China	18 Premenopausal 44 Postmenopausal	24 Premenopausal 46 Postmenopausal	37.06 Premenopausal 57.45 Postmenopausal	35.52 Premenopausal 56.89 Postmenopausal	Explored the differences of microbial gut community and functional capabilities between postmenopausal breast cancer patients and healthy controls using shotgun metagenomic analysis	Case–control study	Feces	Metagenomic sequencing using Illumina HiSeq	Richness: Chao1 Diversity: Shannon index	Jensen-Shannon divergence
Goedert et al. [21]	Colorado, USA	48 (scheduled for treatment of biopsy proven breast cancer)	48	62.17	61.85	Investigated if the gut microbiota of postmenopausal women with incident breast cancer, pre-treatment, differs from control women	Pilot study Case control	Feces	Illumina 300PE MiSeq targeting on V3-V4 region of 16S rRNA gene	Richness: Observed species, Chao1 Diversity: Shannon diversity index, Faith’s PD	Bray Curtis, Unweighted/ weighted UniFrac distance matrices

The overall quality of the eleven non-randomized studies were classified as high-quality studies as the overall score ranges from 60 to 100% (Table 2). Nevertheless, the intervention administered or exposure of the occurred methodology quality criteria is not relevant to our research interest hence not applicable (NA).

Table 2.

Quality assessment using the Mixed Method Appraisal Tool (MMAT) for quantitative non-randomized studies

References	Type of study	Screening questions		Methodology quality criteria					Overall quality score (%)
		Are there clear research questions? (%)	Do the collected data allow to address the research questions? (%)	Are the participants representative of the target population? (%)	Are measurements appropriate regarding both the outcome and intervention (or exposure)? (%)	Are there complete outcome data? (%)	Are the confounders accounted for in the design and analysis? (%)	During the study period, is the intervention administered (or exposure occurred) as intended?
Jiang et al. [30]	Case–control study	100	100	60	80	100	100	NA	80
Ma et al. [38]	Case–control study	100	100	60	80	100	100	NA	80
Wu et al. [64]	Case–control study	100	100	100	80	100	100	NA	80
Yao et al. [65]	Case–control study	100	100	50	80	100	100	NA	60
Byrd et al. [9]	Population based Case–control study	100	100	100	80	100	100	NA	80
He et al. [25]	Case–control study	100	100	100	80	80	100	NA	80
Hou et al. [27]	Case–control study	100	100	80	80	80	100	NA	80
Bobin-Dubigeon et al. [7]	Case–control study	100	100	60	80	80	100	NA	80
Wu et al. [63]	Case study	100	100	50	80	100	100	NA	60
Zhu et al. [67]	Case–control study	100	100	80	80	100	100	NA	80
Goedert et al. [21]	Pilot study Case control	100	100	80	80	100	100	NA	80

The score was presented as 100% quality criteria met; 80% quality criteria met; 60% quality criteria met; 40% quality criteria met; 20% quality criteria met

Fecal Microbiota Composition Significance in Association with Breast Cancer

All 11 of the included studies had sequenced the gut microbiota through the 16S rRNA gene of breast cancer cases and control to study the alpha- and beta-diversity of gut microbiome between both groups. However, beta-diversity results for studies Jiang et al. [30] and Bobin-Dubigeon et al. [7] was not included as the studies did not provide the analysis or did not report the exact values obtained. Additionally, Byrd et al. [9] analyzed the association of gut microbiome with breast cancer using a conventional method by not sorting feces microbiota to the intestinal secretory immunoglobulin A (IgA) goups like IgA-positive and IgA-negative clustered.

The combined results of these studies were depicted in Table 3. In terms of species richness assessed through Observed ASVs, Observed species, Pielou’s evenness as well as Chao1, constant statistical differences were observed across all study cases except for a few studies such as those conducted by Bobin-Dubigeon et al. [7], He et al. [25], Yao et al. [65] and Zhu et al. [67]. Similarly, the assessment of species diversity through Shannon, Faith’s phylogenetic, and Simpson diversity indices returned mostly insignificant statistical results with a few exceptions. These were the studies conducted by Jiang et al. [30], Ma et al. [38], Wu et al. [64], the case control study conducted in Ghana by Byrd et al. [9] as well as the specific results which compared Pre-breast cancer groups against Pre-control groups by Hou et al. [27].

Table 3.

Statistical analysis of alpha- and beta-diversities assessment of reviewed studies

References	Alpha diversity		Beta diversity
Jiang et al. [30]	Significant alpha diversity results where species richness was higher within breast cancer patients vs healthy control	Ace index, (P ≤ 0.05) Chao index, (P ≤ 0.05) Shannon index, (P ≤ 0.05)	Samples were observed to not cluster together, suggesting different compositions	NA
Ma et al. [38]	Alpha diversity was much lower within breast cancer patients compared to healthy control	Observed ASVs, (P ≤ 0.05) Chao1, (P ≤ 0.05)	Significant differences were observed across all three groups (breast cancer, benign breast lesion, healthy control) in terms of compositions	Weighted Unifrac distance, (P ≤ 0.05) Unweighted Unifrac distance, (P ≤ 0.05)
	No significant difference in alpha diversity between healthy individuals and patients with benign breast lesions	Observed ASVs, (P ≥ 0.05) Chao1, (P ≥ 0.05)
Wu et al. [64]	Significant alpha diversity results between breast cancer cases against healthy control	Observed ASV, (P ≤ 0.05) Faith’s PD, (P ≤ 0.05) Shannon index, (P ≤ 0.05)	Siginificant differences for beta diversity between breast cancer cases and non-malignant cases against healthy control	Bray Curtis, (P ≤ 0.01) Jaccard, (P ≤ 0.01) Weighted Unifrac distance, (P ≤ 0.01) Unweighted UniFrac distance, (P ≤ 0.01)
	Significant alpha diversity results between non-malignant cases against healthy control	Observed ASV, (P ≤ 0.05) Faith’s PD, (P ≤ 0.05) Shannon index, (P ≤ 0.05)
	No significant alpha diversity results between non-malignant cases against breast cancer cases	Observed ASV, (P ≥ 0.05) Faith’s PD, (P ≥ 0.05) Shannon index, (P ≥ 0.05)	No significant differences for beta-diversity between non-malignant cases and breast cancer cases	Bray Curtis, (P ≥ 0.01) Jaccard, (P ≥ 0.01) Weighted Unifrac distance, (P ≥ 0.01) Unweighted UniFrac distance, (P ≥ 0.01)
Yao et al. [65]	No significant differences for alpha diversity between sleep disturbance group and no sleep disturbance group	Observed ASVs, (P ≥ 0.05) Shannon index, (P ≥ 0.05) Simpson index, (P ≥ 0.05)	Significant beta diversity result between the two groups	Weighted UniFrac distance, (P ≤ 0.05)
Byrd et al. [9]	Breast cancer cases compared with control had statistically lower alpha diversity	Shannon diversity index (P ≤ 0.05) Observed ASVs (P ≤ 0.05) Faith’s PD (P ≤ 0.05)	Significant different in Beta diversity of fecal microbiota in breast cancer and non-malignant breast disease, compared to control	Bray Curtis, (P ≤ 0.001) Weighted UniFrac distance matrices, (P ≤ 0.001) Unweighted UniFrac distance matrices, (P ≤ 0.001)
	No significant different of alpha diversity index between non-malignant cases vs breast cancer cases	Shannon diversity index, (P ≥ 0.05) Observed ASVs, (P ≥ 0.05) Faith’s PD (P ≥ 0.05)	No significant different of Beta diversity of fecal microbiota in breast cancer and nonmalignant breast disease	Bray Curtis, (P ≥ 0.05) Weighted UniFrac distance, (P ≥ 0.05) Unweighted UniFrac distance, (P ≥ 0.05)
	Non-malignant cases compared with control had statistically lower alpha diversity	Observed ASVs, Faith’s PD (P ≤ 0.05) Except Shannon index (P ≥ 0.05)
He et al. [25]	Breast cancer patients showed no statistical differences to control group; both are similar in number of intestinal flora	Pielou’s evenness, (P ≥ 0.05) Shannon index, (P ≥ 0.05) Simpson index, (P ≥ 0.05)	Significant separation showing composition of intestinal flora differed significantly between control and breast cancer group	Redundancy Analysis (RDA), (P ≤ 0.05)
Hou et al. [27]	Alpha-diversity specifically decreased within premenopausal breast cancer patients while no differences were spotted between both groups for who are postmenopausal in status	Shannon index; Post-control and post-breast cancer (P ≥ 0.05) Pre-breast cancer and post-breast cancer (P ≥ 0.05) Pre-breast cancer and Pre-control (P ≤ 0.05)	Significant difference in microbial composition observed between groups	Unifrac distance, (P ≤ 0.05)
Bobin-Dubigeon et al. [7]	No statistical differences for Chao1 metric but Shannon index showed lower diversity within cancer group vs control group	Rarefied Chao1, (P ≥ 0.05) Shannon index, (P ≥ 0.05)	Data not shown	NA
Wu et al. [63]	Significant differences observed for alpha diversity between HER + and HER- groups. Other factors showed no significant findings	Observed ASVs, (P ≤ 0.05) Shannon index, (P ≤ 0.05) Chao1, (P ≤ 0.05)	Significant difference for beta diversity based on Unweighted UniFrac distance	Unweighted UniFrac distance, (P ≤ 0.05)
			No significant difference based on Weighted Unifrac	Weighted UniFrac distance, (P ≥ 0.05)
Zhu et al. [67]	No significant alpha diversity results based on comparison between premenopausal breast cancer patients and premenopausal healthy control	Chao1, (P ≥ 0.05) Shannon index, (P ≥ 0.05)	No significant difference in composition between premenopausal breast cancer patients and premenopausal healthy control	Jensen-Shannon divergence, (P ≥ 0.05)
	Significant alpha diversity results based on comparison between postmenopausal breast cancer patients and postmenopausal healthy control	Chao1, (P ≤ 0.05) Shannon index, (P ≤ 0.05)	Significant difference in composition between postmenopausal breast cancer patients and postmenopausal healthy control	Jensen-Shannon divergence, (P ≤ 0.01)
Goedert et al. [21]	Breast cancer patients compared with control, had statistically lower alpha diversity	Observed species, (P ≤ 0.05) Chao1, (P ≤ 0.05) PD_whole tree, (P ≤ 0.05) Except Shannon, index (P ≥ 0.05)	Significant different in Beta diversity	Bray Curtis, (P ≥ 0.05) Weighted UniFra,c distance (P ≥ 0.05), Unweighted UniFrac distance, (P ≤ 0.05)

Captivatingly, while Goedert et al. [21] reported a significant difference in beta-diversity of breast cancer cases compared to control, this result is only based on p-value of Unweighted UniFrac distance matrices. Alternatively, all other studies reported similar findings,statistically significant results for beta-diversity analysis, suggesting that the composition of gut microbiome differs between breast cancer patients and the control group. The study under Bobin-Dubigeon et al. [7] however, did not show their results as said data was not made available for the public under ethical reasoning. Similarly, Jiang et al. [30] did not share the obtained statistical values obtained, instead only reporting significant differences based on figure only. Overall, by combining these studies, the fecal microbiota of breast cancer cases were observed to have generally lower statisical alpha-diversity values and showed different intestinal gut flora compositions versus the control group.

Significant Taxa Associated with Breast Cancer

Detailed reports regarding the analyzed taxa association in respect to the eleven studies can be found within Supplementary Table 1. These highlighted features were cross-referenced with previously known producers of GUSs [34]. The comparisons which were either postively (scored 1) or negatively (scored − 1) correlated with breast cancer cases were summarized in Fig. 2. Interestingly, the asscociations of breast cancer cases with GUSs producing bacteria showed differing patterns in respect to the other studies. Reported gmGUSs which showcased positive correlation are of the following genera: Weissella, Turicibacter, Ruminococcus, Prevotella, Porphyromonas, Megamonas, Marvinbryantia, Klebsiella, Fusobacterium, Faecalibacterium, Eubacterium, Escherichia, Enterococcus, Enterobacter, Coprobacillus, Clostridium, Butyricicoccus, Blautia, Bacteroides, Anaerostipes, Actinomyces, and the order Clostridiales. Negative correlated gmGUSs were found to be Turicibacter, Staphylococcus, Ruminococcus, Propionibacterium, Prevotella, Porphyromonas, Pediococcus, Parabacteroides, Odoribacter, Megamonas, Lactobacillus, Fusobacterium, Faecalibacterium, Eubacterium, Escherichia, Dorea, Collinsella, Butyricicoccus, Blautia, and Bifidobacterium.

Fig. 2.

Highlighted taxa of GUSs bacteria correlated either positively (scored 1) or negatively (scored − 1) with breast cancer cases across each reviewed study

To note there a quite a number of features in which the reviewed studies reported conflicting findings. In the case of the genus Turicibacter, both Geodert et al. [21] and He et al. [25] reported its postive correlations with breast cancer whereas Bryd et al. [9] suggested the opposite. The studies conducted by Hou et al. [27], Wu et al. [63], and Zhu et al. [67] found that genus Ruminoccocus was negatively associated with breast cancer. In contrast, Jiang et al. [30] depicted that the bacteria had positive correlations instead. Additionally, the studies by both Byrd et al. [9] and Wu et al. [64] suggested the genus Prevotella to negatively associated with breast cancer, though the studies of Ma et al. [38], Yao et al. [65] and Zhu et al. [67] reported it to have postive correlations instead. For the genus Porphyromonas, Ma et al. [38], suggested postive correlations while Wu et al. [64] reported negative associations to breast cancer. Similarly, He et al. [25] found the genus Megamonas to be less abundant within breast cancer patients while Ma et al. [38] reported its increased abudance within those same kinds of individuals. Moreover, the genus Fusobacterium was reported to be negatively correlated to breast cancer based on He et al. [25] and Wu et al. [64] but Goedert et al. [21] instead reports it to have postive correlation.

The studies conducted by Bobin-Dubigeon et al. [7],Byrd et al. [9], Goedert et al. [21] and Jiang et al. [30] reported that the genus Faecalibacterium to be positively correlated with breast cancer. In contrast, the study done by Hou et al. [27] showed the opposite in which Faecalibacterium were negatively correlated with breast cancer. Furthermore, the genus Eubacterium was observed to be negatively correlated to breast cancer based on Byrd et al. [9] and Zhu et al. [67] while He et al. [25] suggest it to be postively asociated instead. In the case of the genus Escherichia, Bobin-Dubigeon et al. [7] suggested it to have negative associations with breast cancer while Zhu et al. [67] reported it to possess postive correlations instead. Besides that, He et al. [25] and Jiang et al. [30] reports that the genus Butryricicoccus to be postively correlated with breast cancer while Goedert et al. [21] suggests otherwise. Finally, the study conducted by Wu et al. [63] found the genus Blautia to be less present within breast cancer patients while the study by Bobin-Dubigeon et al. [7] instead suggested it to be more prevalent within those patients.

Discussion

Identification of total microbial community in defined specific environments and compared diversity metrics across different ecological gradients is feasible with the advent of high throughput sequencing technology [24, 37, 53, 54, 60, 62]. After the raw sequencing files were generated, the data would be processed and taxonomy was assigned using 16S rRNA gene database [23, 31], a summary statistic within sample diversity, and estimation of similarity or dissimilarity between samples are analyzed using alpha- and beta-diversity matrices respectively [60]. Alpha- and beta-diversities were considered to be important indices to describe the characteristics of the gut microbiota in association to the occurrence and development of many diseases. It was anticipated that it would decline in disturbed systems, and low diversity is associated with the presence of diseases and ill health, or dysbiosis. However, the relationship between the gut microbiota and breast cancer has not been fully uncovered by a systematic review, and earlier reports were restricted to specific individual study from different regions. This study focused on determining if the gut microbiota's alpha- and beta-diversities may be utilised as a predictor of breast cancer, which is crucial for the early diagnosis of these diseases that substantially impair patients' quality of life.

All studies reviewed found significant lower alpha-diversity in breast cancer cases than control groups. Even though all of the included studies used high throughput sequencing of 16S rRNA targeting either V3, V4 or both V3 and V4 regions, the alpha-diversity matrices used to measure gut microbiome alpha-diversity was diverse. Hagerty [22] listed 18 alpha diversity measures that were calculated in QIIME 2,Ace, Observed species, Chao1, Margalef, Fisher alpha, Faith_PD, Brillouin d, Shannon, Enspie (ENS), Menhinick, Mcintosh e, Simpson, Berger Parker d, Strong, Simpson e, Pielou e, Heip e, and Lladser pe. Nevertheless, a recent new method, Observed ASV, that infer the biological sequences in the sample prior to the introduction of amplification and sequencing errors, and distinguish sequence variants differing by as little as one nucleotide [10], is not listed by Hagerty. Although statistically a richness analysis of three reviewed studies is significant, to reduce bias in alpha-diversity assessment, using multiple matrices in analyzing richness or evenness, or diversity is recommended.

In this study, one of the reviewed literature suggests that there is no significant difference between the alpha- and beta-diversity indexes between non-malignant cases in comparison to breast cancer cases. Unfortunately, an unequal sample size for breast cancer cases which doubles compared to non-malignant cases might be the root cause of the issue with the alpha-diversity analysis [62]. The same can also be said for the studies conducted by Bobin-Dubigeon et al. [7], He et al. [25] and Yao et al. [65] which showed no significant differences for alpha-diversity results,owing to either their uneven sample size comparison or having too small of sample size pool.

In this review, the beta-diversity analysis for the gut microbiome of breast cancer cases showed different outcomes based on their resepctive studies. For a study conducted within the Ghana population, significant differences were observed when compared to the control group, however within a population study of Colorado, USA the unsorted microbiome groups were reported to be not significant however, based on breast cancer cases versus control group it was reported to be significant in terms of beta-diversity. Similarly, other studies reported significant findings for beta-diversity analysis [25, 27, 38, 44, 63–65]. It should also be noted that the beta-diversity results for Bobin-Dubigeon et al. [7] were not available as the data was not shown and was not made available for public viewing due to ethical reasons. Similarly, no statistical values were recorded for the study by Jiang et al. [30], though it was stated that there were significant results. This suggests that although strong population homogeneity was within the strength of the study, different geographic/ethnic dietary across studies contribute to difficulties in comparability of beta diversity analysis. The selection of sample size also plays a crucial role behind the reasoning of such varying results.

The differences in outputs between non phylogenetic measures of community similarity, Bray Curtis, and phylogenetic based Unweighted and Weighted UniFrac distance matrics presented in reviewed studies may lead to questionable significant difference in beta-diversity analysis. As it is, it can be considered to be statistically significant between the cases even if only one method used is significant. Nevertheless, the comparisons from the study of Goedert et al. [21] provide compelling evidence that plot-to-plot dissimilarity abundance of species analyzed through Bray Curtis, suggests that there is no changes of species abundance in breast cancer patients when the number of samples from studied population is not big enough, as compared to study conducted by Byrd et al. [9]. In a similar fashion, while the studies of He et al. [25], Hou et al. [27] and Yao et al. [65] uses only a single metric, the dissimilarity plot produced can still vouch for said results. However, analysis of phylogenetic based Unweighted UniFrac distance matrics of Byrd et al. [9], Goedert et al. [21] , Ma et al. [38], Wu et al. [63] and , Wu et al. [64] indicated a significant difference for the beta-diversity analysis which suggests, that qualitative measure of beta-diversity between normal and breast cancer cases depending on the presence or absence of the taxa are different from each other. One study confirmed that UniFrac values can be influenced by the number of sequences or samples [36]. Based on the reviewed studies, only Byrd et al. [9] shows significant differences in the observed beta-diversity analysis for unsorted microbiota and IgA negative microbiota cases in using the Weighted (quantitative) UniFrac distance method. One study suggests that, beta-diversity tests using unweighted and weighted UniFrac distances are less powerful in detecting abundance change in moderately abundant lineages, thus suggesting the generalized UniFrac distance is most powerful in detecting rare and highly abundant lineages [13].

At a cursory glance, there appears to be no clear link between breast cancer pathogenesis and the gut microbiome. However, there a few current theories which debunk such statements. The most promising thought of cancer development which is specific for breast cancer pathogenesis is the estrobolome concept. The term ‘estrobolome’ was established in the year 2011 and refers to the collective enteric bacteria which are capable of metabolizing estrogen [45]. Past studies have shown that the estrobolome may play a key role in some of the hormonal disorders affecting women such as endometrial, ovarian and breast cancers [14, 45]. This is likely due to the effect of estrogen which may stimulate the growth of cells within these specified regions. One the mechanisms associated with the estrobolome is the presence of bacterial GUSs,more accurately the term estrobolome refers to the bacteria capable of synthesizing GUSs [45]. Estrogens are mostly produced within the adrenal glands, ovaries and adipose tissue which may then either circulate throughout the body freely or in a protein-bound form. Regardless, these protein molecules would first be metabolized in the liver which will conjugate both estrogens and their respective metabolites. Ordinarily, conjugated estrogens leave the body in the form of urine or bile in feces through metabolic conversion to water-based molecules. However, GUSs posses the capability to deconjugate these estrogen molecules which not only makes them more biologically active but also more suceptible to be reabsorbed and circulate throughout the body [34, 55]. As such this enables them to bind to estrogen receptors which activates and increases the number of G0/G1 cells which enter the cell cycle and promote cell proliferation [5]. In the case of gut dysbiosis then, major shifts to the bacteria compostion could result in the increase of gmGUSs, thus ultimately increasing the overall estrogen which circulates the body. Interestingly, a shift which result in the decrease of GUSs producing bacteria would suggest the opposite then; a lack of circulating estrogen which may bring about hypoestrogen related issues instead [5].

Interestingly, among the reviewed studies, attempts were made to link estrogen associations to the gut microbiome. Despite the two-fold higher levels of estrogen metabolites among post-menopausal women with breast cancer, and significant changes of alpha- and beta-diversities of gut microbiomes in breast cancer cases, Goedert et al. [21] reported that gut microbiome are not correlated with the increase of estrogen metabolites level within breast cancer cases, thus providing no support to the hypothesis that the microbiota substantially metabolises estrogens. While this outcome is unexpected, this could be due to small number of sample sizes used, of which both studies used 48 breast cancer cases [21]. A case control clinical trial on association of breast and gut microbiota dysbiosis and the risk of breast cancer in which was conducted in Spain suggested the sample size should be more than 100 women with breast cancer, matched with 100 control women subjected to breast reduction or augmentation, of which calculated based on consideration of the incidence of breast cancer in Spain, the female population of Andalusia and the formula described by Aguilar-Barojas [44].

Known mgGUSs associated with the reviewed studies were reported in the results section. As noted, the features were categorized to be either postively or negatively correlated in regards to their respective studies. While positive correlation implies that the bacterial content increases within breast cancer patients, negative correlation instead suggest that the bacteria count decreases within the target group. Nevertheless, this should not discredit completely the involvement of bacteria which show negative correlation. Statistically speaking, even negatively correlated variables may still play a role in prediction as compared to variables with zero correlation [16]. As such, the presence of these mentioned bacteria are only diminished at best and are still quite present within the obseverd gut microbiome. Therefore, given the capabilities of these bacterial features, regardless of their correlation status to the patients, it can be speculated that such bacteria may increase the overall amount of estrogen circulating within the body as explained previously. This spike of estrogen may promote cell growth which may spiral out of control and create tumors within the breast region of the patients. The endogenous estrogen and estrogen-derived metabolite thus, are strongly correlated with breast cancer risk in postmenopausal women [33, 49]. Hence, a gut microbiome experiencing dysbiosis at long-term may greatly increase the risk for the development of breast cancer [6, 11, 12, 17–19, 35, 39]. An atlas of gmGUS from 139 individuals faecal samples revealed 4 phyla observed to express the GUS protein includes; Bacteroidetes, Firmicutes, Verrucomicrobia and Proteobacteria [46]. Of these, Bacteroidetes phylum exhibits the broadest diversity and expressed largest number non-redundant GUS [46]. This regulatory nature of the estrogen cycle and GUSs might be the missing link which connects breast cancer development and the gut microbiome.

Besides that, other factors may also contribute to the development of breast cancer within individuals. One of the most prevalent theories include the leaky gut theory which proposes that harmful pathogens situated in the gut escapes out and causes issues to arrise in other region of the body. The majority of immulogical active cells are prevalent within the gastrointestinal tract tissues which has the highest presence of immunogenic agents such as the components of the microbiota and food [48]. Ordinarily, epithelial cells form a layer of protection and have evolved over time to prevent the invasion of such foreign bodies. One of these protection mechanisms exists in the form of mucin or mucos layer which often serves to separate commensal bacteria as well as their components from the epithelial layer. The gut microbiome play a role in the maintenance of this mucosal barrier as well as determining the permeabilty of the epithelial wall [15, 48]. When the gut experiences dysbiosis, that is the a sudden shift in microbiota composition, it interupts the stability which is maintaining the safety net established. This in turns reduces the mucosal layer as well in a sense increases the permeabilty of the epithelial wall to pathogen invaders. Commensal and pathogenic bacteria would then be able to pass through the wall and travel to other regions of the body through its circulation system. Inflammatory agents produced from said bacteria would also spread in a similar fashion which may intiate tumorigenesis within specific site locations upon their arrival to that locale. Similarly, the bacterial metabolites which have invaded the body may also cause a region to experience hypercoagulation, a state which supports as well as contributes to the growth and development of tumor cells [15]. As such it can be said, shifts towards the gut microbiome may trigger the ‘leaking’ of certain bacteria as well as metabolites into other regions of the body where they may initiate inflammation to occur which ultimately brings rise to tumor formation.

To the best of our knowledge, this is the first systematic review that utilizes diversity indexes in determining gut microbiome markers correlated with breast cancer. Detailed analysis on each taxon in relation to their capabilities to metabolize estrogen and correlation status to breast cases provided within this review may give insights into possible early detection methods for breast cancer development. By excluding external factors such as special diet, and exercise the possible bias in interpretating the results were able to be reduced in this review. Through the selection of literature which exclusively involves the use of stool specimens that were collected prior to any treatment, a fair review on the effects of gut microbiome dysbiosis in breast cancer in comparison to control groups were able to be conducted.

Our review of this literature is limited to a descriptive approach and no meta-analysis was conducted. Because these studies had different aims, studied heterogenous populations, reported on relative rather than absolute abundance of each taxon that may correlate with breast cancer, hence a meta-analysis approach would not be suitable for this review. The search for gut microbiota via feces excluding breast microbiota carries specific limitations on the association of microbiota in breast cancer cases. We further note that breast microbiota may be equally or more relevant in association of gut microbiome with breast cancer [17]. We also acknowledge that most of the reviewed studies do not highlight GUSs associations in their original work and thus may not address them thoroughly. Besides, many of these early findings are merely correlative and associative, but lacking in describing the real mechanisms in shaping the phenotypes. It must be stressed upon that correlation does not imply causation, this review aims to aid by directing or narrow down and improve studies to understand metabolism and look for possible causes of breast cancer pathogenesis through the lens of the gut microbiome. This study only included studies published in English and may miss publications in other languages. However, this limitation is due to reviewers’ limitation in understanding languages other than English.

Conclusions

In conclusion, the likelihood of gut microbiomes having an effect towards breast cancer development is quite high. While it may not be the sole reason for cancer pathogenesis within the breast, it still represent a possible risk factor for future diagnosis. Also worth noting is that bigger sample sizes are critically important to draw a significant contribution of gut microbiome dysbiosis on endogenous estrogen and estrogen metabolites, hence breast cancer prognosis. Larger studies may employ a functional system of categorization that may include type of diets depending on geographical location that may be important covariates. Stressing on the finding of significant taxa at species level may prove useful in better understanding the contribution of gut microbiome as a biomarker to breast cancer prognosis. This is made more clear in identifying the presence of gut bacteria producing GUSs which play a large role within the estrobolome. Moving forward, the data from this study is important in order to realise the gut microbiome’s potential in health and disease, for which a comprehensive study concentrating on establishing causality and molecular mechanism could be designed. As it stands, there exist a very clear need to further investigate the interaction effect between the estrobolome and the gut microbiome for their role in the development of breast cancer within individuals.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

ASMY, NEM, and NSK performed literature review. ASMY and NSK wrote the first draft of the manuscript. HFA, NEM, NSA, DDT, and NMM critically reviewed the manuscript with respect to their areas of expertise. NEM, HFA, and NSA conceptualized the project. SAT, NEM, DDT, NMM and HFA critically reviewed the statistical analysis. All authors contributed to writing and revising the manuscript.

Funding

This research received funding from the Ministry of Higher Education, Malaysia under the grant RACER/1/2019/SKK08/UIAM//1, and Universiti Malaysia Pahang, Malaysia for providing additional support through the grant PGRS220383.

Data Availability

Not applicable.

Declarations

Conflict of interest

The authors declare no conflict of interest.