🧬 biology

Causal Relationships between Gut Microbiota, Immune Cells, and Breast Cancer: A Two-step, Two-sample Mendelian Randomization Study

This two-step, two-sample Mendelian randomization study provides genetically supported evidence that specific gut microbiota, such as *Rhodanobacter*, causally increase breast cancer risk partially through the mediation of specific immune cell characteristics, notably CD25hi CD45RA+ CD4 non-Treg cells.

Original authors: Haiquan Liu, Weiwen Li, Mingtao Shao

Published 2026-07-21

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Haiquan Liu, Weiwen Li, Mingtao Shao

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine your body as a bustling, high-tech city. In this city, there's a massive recycling plant called the gut, filled with trillions of tiny workers known as gut microbiota. These workers aren't just cleaning up; they're constantly sending messages to the city's security force, the immune system. Sometimes, these messages help the security force stay sharp, and sometimes, they accidentally give the wrong orders. Now, picture a specific type of troublemaker in the city: breast cancer. Scientists have long suspected that the workers in the recycling plant might be whispering secrets to the security guards that help the troublemakers grow, but it's been hard to prove who is talking to whom and who is actually in charge. Is the gut causing the cancer, or is the cancer changing the gut? To solve this mystery without waiting decades for a real-life experiment, researchers use a clever detective trick called Mendelian Randomization. Think of this as using your DNA as a "genetic lottery ticket." Since you inherit your genes at birth, before any disease starts, they act like a frozen snapshot of your gut bacteria and immune system. By checking these genetic tickets in thousands of people, scientists can figure out if a specific gut worker is truly the boss of the cancer, rather than just a bystander.

This study, led by researchers at Jiangmen Central Hospital, decided to play detective with a massive amount of genetic data to map out the exact chain of command between gut bacteria, immune cells, and breast cancer. They looked at data from nearly 6,000 people to see which gut bacteria were linked to specific immune cells, and then checked if those same bacteria and immune cells were linked to breast cancer risk in over 200,000 people. Their goal was to see if the gut bacteria were directly causing the cancer, or if they were first messing with the immune system, which then allowed the cancer to grow.

The investigation revealed a fascinating, albeit partial, story. The researchers found that out of hundreds of gut bacteria types, five specific groups were positively linked to a higher risk of breast cancer. One of these troublemakers, a bacterium called Rhodanobacter, stood out as a key suspect. But here is the twist: Rhodanobacter doesn't seem to attack the city directly. Instead, it appears to send a signal to a specific type of security guard in the immune system. The study identified two specific groups of immune cells, named CD25hi CD45RA+ CD4 not Treg %CD4+ and CD25hi CD45RA+ CD4 not Treg %T cell, which act as the middlemen.

The data suggests that having more Rhodanobacter causes an increase in these specific immune cells. In turn, having more of these immune cells is linked to a higher risk of breast cancer. The researchers calculated that this "gut-to-immune-to-cancer" pathway explains about 4.64% to 5.02% of the total risk associated with Rhodanobacter. In other words, the gut bacteria is like a troublemaker who hires a specific security guard to do its dirty work, but it's not the only way the troublemaker causes chaos.

The study also checked the reverse direction to make sure they weren't mixing up cause and effect. They asked: "Does having breast cancer change the amount of Rhodanobacter in the gut?" The answer was a clear no. The cancer doesn't seem to be the one creating the bacteria; the bacteria seems to be influencing the cancer risk. Similarly, they found that 13 other types of gut bacteria might actually be good guys, potentially lowering the risk of breast cancer, while 39 different types of immune cells were linked to the disease, with 17 of them increasing the risk.

However, the researchers are careful not to claim they have solved the whole puzzle. They point out that while they found a clear link, the immune cells they identified only explain a small slice (less than 5%) of the total effect. This means there are likely other secret pathways and middlemen they haven't found yet. Also, because all the data came from people of European descent, we don't know if these same rules apply to people from other parts of the world. While the study provides strong genetic evidence that Rhodanobacter can nudge the immune system toward a state that favors breast cancer, it suggests this is just one piece of a very complex, multi-layered mystery that scientists are still working to fully understand.

Technical Summary: Causal Relationships between Gut Microbiota, Immune Cells, and Breast Cancer

Problem Statement
While prior research has established associations between gut microbiota (GM) and breast cancer (BC), the underlying causal mechanisms and mediating pathways remain insufficiently elucidated. Specifically, the extent to which GM influences BC risk directly versus through the modulation of immune cell (IC) populations is not fully understood. This study addresses this gap by investigating the causal relationships between 473 GM taxa and BC risk, while simultaneously exploring the mediating role of 731 distinct immune cell characteristics.

Methodology
The study employed a two-step, two-sample Mendelian randomization (MR) framework using publicly available Genome-Wide Association Study (GWAS) data. The analysis followed a three-step design:

GM to BC: Determining the causal effect of GM taxa on BC risk.
IC to BC: Evaluating the causal effect of IC traits on BC risk.
Mediation Analysis: Examining the causal impact of GM on specific IC traits and quantifying the proportion of the GM-BC causal effect mediated by these ICs.

Data Sources and Instrument Selection:

Exposure (GM): Data from the FINRISK 2002 cohort (5,959 individuals), encompassing 473 taxa across 11 phyla.
Mediator (IC): Data covering 731 immune cell characteristics (including B-cells, dendritic cells, T-cells, monocytes, myeloid cells, TBNK, and Treg cells) from a GWAS of 731 traits.
Outcome (BC): Data from the FinnGen R10 database (201,713 individuals; 18,786 cases, 182,927 controls).
Instrumental Variables (IVs): SNPs were selected with a genome-wide significance threshold of $P < 1 \times 10^{-5}$ for GM and IC traits, and $P < 5 \times 10^{-8}$ for BC. SNPs were clumped ( $R^2 < 0.001$ ) to ensure independence, and F-statistics $> 10$ were required to mitigate weak instrument bias. Potential pleiotropy was assessed via PhenoScanner V2, MR-Egger intercept tests, and MR-PRESSO.
Statistical Analysis: The Inverse Variance Weighted (IVW) method served as the primary analysis, supplemented by MR-Egger, weighted median, and mode methods. Sensitivity analyses included Cochran's Q for heterogeneity and leave-one-out (LOO) analysis.

Key Contributions and Results

Causal Impact of GM on BC:
The analysis identified 18 GM taxa with significant causal associations with BC.
- Protective Effects (13 taxa): Including CAG-390 sp003523225, Akkermansia muciniphila B, Ruminococcus, and Achromobacter, among others.
- Risk-Increasing Effects (5 taxa): Clostridium tertium, Prevotellamassilia, Brachyspirales, Providencia, and Rhodanobacter. Specifically, Rhodanobacter was associated with an increased BC risk (IVW: OR = 1.3665, 95% CI: 1.0618–1.7586, $P = 0.0153$ ).
Causal Impact of Immune Cells on BC:
The study identified 39 IC characteristics linked to BC risk, with 17 demonstrating an increased risk and 22 showing protective associations.
Mediation Analysis (The Tripartite Interaction):
A critical finding was the identification of a specific mediating pathway. The causal effect of Rhodanobacter on BC was found to be partially mediated by two specific immune cell phenotypes:
1. CD25hi CD45RA+ CD4 not Treg %CD4+
2. CD25hi CD45RA+ CD4 not Treg %T cell
The mediation analysis revealed:
- Rhodanobacter significantly increased the levels of these specific IC traits (IVW $P = 0.0043$ and $P = 0.0116$ , respectively).
- These IC traits were causally associated with increased BC risk.
- Mediation Proportion: The indirect effect of Rhodanobacter on BC via CD25hi CD45RA+ CD4 not Treg %CD4+ accounted for 4.64% of the total causal effect. The effect via CD25hi CD45RA+ CD4 not Treg %T cell accounted for 5.02%.
Reverse MR:
Bidirectional MR analyses confirmed no significant reverse causality (i.e., BC did not causally influence Rhodanobacter or the identified IC traits), reinforcing the directionality of the findings.

Significance and Claims
The paper claims to provide genetically supported evidence linking gut microbiota, immune modulation, and breast cancer etiology. By utilizing MR, the study offers robust validation of a tripartite interaction where specific gut bacteria (Rhodanobacter) influence breast cancer risk not only directly but also through the modulation of specific T-cell subsets (CD25hi CD45RA+ CD4 non-Treg cells).

The authors position these findings as a step toward understanding the complex biological mechanisms driving BC initiation and progression. They emphasize that while the mediation effects (approx. 4.6–5.0%) are partial, they establish a concrete causal pathway involving the gut-immune-tumor axis. The study concludes that these results highlight the potential causal link between GM dynamics, immune cell profiles, and BC, offering a foundation for future research into immune-mediated therapeutic strategies.

Limitations Noted by the Authors
The authors explicitly state several limitations:

Univariable MR: The study could not fully disentangle independent effects from correlated risk factors (e.g., obesity) without multivariable MR (MVMR).
Population Specificity: All data were derived from European populations, limiting generalizability to other ethnic groups.
Mediation Scope: The identified mediation effects account for a small proportion of the total effect, suggesting other unmeasured mediating factors exist.
Data Source: Reliance on public databases meant environmental factors were not directly controlled, though genetic instruments were used to mitigate confounding.

Technical Summary: Causal Relationships between Gut Microbiota, Immune Cells, and Breast Cancer

More like this