Tumor-infiltrating microbes and therapy response: a new frontier in triple-negative breast cancer precision oncology.

Microbial biomarkers in TNBC
The clinical landscape of breast cancer is rapidly evolving, with growing recognition of tumor-infiltrating microbes’ potential as a prognostic and predictive biomarker. In this issue, Chen et al add to our understanding by demonstrating how tumor-infiltrating microbes can predict pathologic complete response (pCR) following chemotherapy plus immunotherapy treatment in triple-negative breast cancer (TNBC). This growing field offers new opportunities to identify novel therapeutic correlates and refine risk stratification across breast cancer subtypes.
TNBC remains a challenging subtype due to its aggressive biology and limited targeted options. The landmark KEYNOTE-522 trial demonstrated that adding pembrolizumab to neoadjuvant chemotherapy significantly improved pCR rates and event-free survival, establishing this regimen as the standard of care for operable TNBC, particularly in T2N0 or node-positive disease.1
Despite these advances, significant heterogeneity in treatment response persists, underscoring the need for better predictive biomarkers. Increasing evidence suggests that tumor-infiltrating microbes may influence both the efficacy of chemotherapy2 and immune activation3 (figure 1). Characterizing these microbial signatures could help explain differential responses and identify novel resistance mechanisms, particularly in TNBC, where programmed death-ligand 1 (PD-L1) quantity alone is insufficient for patient selection.
Furthermore, the clinical importance of tumor-infiltrating microbes extends beyond TNBC, with promising data implicating its role as a prognostic and predictive biomarker in hormone receptor-positive and HER2-enriched breast cancers.4 Integrating microbial profiling with existing genomic and immune biomarkers could enable more personalized therapeutic strategies, ultimately optimizing outcomes for patients across the spectrum of breast cancer subtypes.

Methodological strengths and interpretive constraints
Chen et al advance our understanding of tumor-infiltrating microbes in TNBC, a technically challenging and emerging field. A key strength is the authors’ multimodal approach, combining 16S rDNA sequencing, single-cell RNA sequencing (scRNA-seq), fluorescence in situ hybridization, and immunohistochemistry. This redundancy addresses major difficulties in the field, including contamination and signal specificity, thereby enhancing the robustness of their findings. Notably, the authors cultivated live microbes from fresh TNBC specimens, enabling orthogonal validation of the 16S results and confirming microbial viability and activity. Finally, the study links distinct microbial signatures to responses to neoadjuvant chemotherapy and immunotherapy, suggesting their potential as predictive biomarkers.
While the study offers valuable insights, several considerations remain. The absence of a model system or experimental controls limits conclusions about causality and generalizability. The scRNA-seq dataset, while informative, is relatively small and supports preliminary findings with modest effect sizes. Additionally, the study does not incorporate established clinical biomarkers such as PD-L1 expression or tumor-infiltrating lymphocytes, which could help contextualize the predictive value of the microbial signatures. The inclusion of a chemotherapy-only comparator group is promising, though a more detailed analysis could further clarify immunotherapy-specific effects. Looking ahead, future studies would benefit from larger cohorts, mechanistic validation, and alignment with clinical standards to more fully define the role of tumor-infiltrating microbes in TNBC treatment response.

Field-wide challenges in microbial profiling
Tumor-infiltrating microbes are an emerging and intriguing aspect of cancer biology. While the presence of microbes in tumors has been recognized in specific contexts, such as human papillomavirus in cervical cancer, recent studies have expanded this understanding to include bacteria and other microbes in solid tumors, including breast cancer.4 5 These microbes may influence cancer development, progression, metastasis, and treatment response, although the mechanisms remain poorly understood.
Recent research has suggested cancer-type-specific microbial signatures using host tumor sequencing data. For instance, Galeano Niño et al6 identified distinct microbial communities in different cancers, including spatial heterogeneity within tumors. However, these findings have met with skepticism due to methodological obstacles. Giwahi et al7 highlighted the risk of false positives in microbiome detection from host sequencing data, noting issues such as contamination, human read removal, and reference genome alignment.
The study of tumor-infiltrating microbes is complicated by the low abundance of microbial signals in human-dominated samples and the technical limitations of current detection methods. A variety of approaches has been employed to detect these microbes, including microscopy, nucleic acid sequencing, and culturing. Microscopy techniques, such as RNAscope, offer high specificity and sensitivity by targeting microbial RNA with fluorescent probes. Culturing remains the gold standard for confirming microbial viability; however, many tumor-associated microbes are challenging to grow under laboratory conditions, and some may be intracellular5 and have lost their exopolysaccharide layer and may not survive the culturing process. Nucleic acid-based methods, including amplicon sequencing and transcriptomics, provide broader detection capabilities but are susceptible to contamination and require careful interpretation. Spatial transcriptomics, in particular, has bridged the gap between sequencing and imaging, enabling researchers to map the presence of microbes within tumor architecture. This method, along with scRNA-seq using microbial-targeted primers, has revealed localized microbial populations and their associations with specific human cell types.8 Experimental models in mice have further validated these findings. For example, microbes introduced via oral gavage have been recovered from tumors, and some have been shown to modulate immune responses or travel with metastases, suggesting functional roles in tumor biology.

Unresolved questions and emerging mechanisms
Despite these advances, fundamental questions remain. The origin of tumor-infiltrating microbes—whether from the gut, oral cavity, bloodstream, or external environment—is still unclear. Transient bacteremias and microbial translocation during homeostasis may contribute to their presence in tumors.9 Moreover, the tumor microenvironment (TME) likely influences microbial persistence, with immune clearance playing a key role. Some microbes may even exploit tumor-specific chemotactic signals to accumulate in these tissues, such as with Salmonella and Escherichia species.
The mechanisms by which tumor-infiltrating microbes influence treatment outcomes are incompletely understood; yet, emerging evidence suggests they can profoundly shape therapeutic efficacy through diverse biochemical and immunological pathways. Microbes possess extraordinary genetic and metabolic diversity, enabling them to modify a wide array of molecular structures, including drugs and host-derived metabolites. Within the TME, this metabolic flexibility allows microbes to degrade or transform therapeutic compounds, particularly those resembling natural products, potentially rendering them inactive. For instance, microbial production of cytidine deaminase has been shown to inactivate the chemotherapeutic agent gemcitabine, an effect reversible by antibiotic treatment2 (figure 1). Similarly, several bacterial strains identified in soft tissue sarcomas were found to degrade doxorubicin. When these doxorubicin-degrading microbes were introduced into mouse models bearing doxorubicin-sensitive tumors, the tumors became resistant to chemotherapy.10 Given that doxorubicin is a naturally derived anthracycline produced by Streptomycetaceae, and gemcitabine structurally mimics natural nucleosides, it is plausible that environmental exposure has driven the evolution of microbial drug-inactivating mechanisms. These findings underscore the potential for tumor-infiltrating microbes to directly interfere with chemotherapeutic efficacy.
Beyond direct drug metabolism, microbes also generate small molecules that modulate immune responses. These microbial metabolites can be either immunostimulatory or immunosuppressive, depending on the context. In melanoma, Lactobacillus reuteri metabolizes dietary tryptophan into indole-3-aldehyde, which activates the aryl hydrocarbon receptor (AhR) in CD8+ T cells, enhancing antitumor immunity and improving responses to immune checkpoint inhibitors.3 Conversely, in pancreatic ductal adenocarcinoma, similar tryptophan-derived indoles activate AhR in tumor-associated macrophages, promoting an immunosuppressive phenotype and dampening antitumor responses.11 This duality underscores the context-dependent nature of microbial metabolite signaling.
Recent studies have also highlighted the ability of specific microbial species or products to modulate immune checkpoint blockade (ICB) responses. For example, Staphylococcus aureus produces α-hemolysin, a pore-forming toxin that increases CD8+ T-cell infiltration and PD-L1 expression in TNBC models, thereby enhancing the efficacy of ICB.12 Similarly, the intratumoral presence of Escherichia coli has been associated with improved overall survival in patients with non-small cell lung cancer receiving ICB, suggesting a beneficial role for certain bacterial taxa in shaping antitumor immunity.13 Additionally, microbes can produce metabolites like inosine, which directly stimulate T-cell activity and enhance the efficacy of ICB.14
Breast tumors express Dectin-1, a C-type lectin receptor that recognizes β-glucans in fungal cell walls. High Dectin-1 expression correlates with poor radiation therapy outcomes, highlighting a potential axis of fungal sensing and immune suppression within the TME.15 Collectively, these findings suggest that it is not merely the presence of specific microbial taxa that matters, but rather the functional roles these microbes play within the TME. We propose that the balance between protumor and antitumor immunity is dictated by the spatial and cellular context of the TME. The immunological outcome is therefore not determined by the microbial product itself, but by which cell types it engages within a given tumor niche. Because different tumor types harbor distinct cellular landscapes, a microbial metabolite or toxin will act on whichever cells are spatially accessible and express the relevant receptor. For example, tryptophan-derived metabolites activating AhR have been shown to drive immunosuppression via macrophages in pancreatic cancer11 but enhance antitumor immunity via CD8+T cells in melanoma.3

Conclusion
Tumor-infiltrating microbes represent a complex ecosystem with implications for cancer diagnosis, prognosis, and therapy. Chen et al link microbial signatures to treatment response in TNBC, underscoring their potential as predictive biomarkers in precision oncology. Key challenges remain: standardizing detection, preventing contamination, and validating microbial activity. Progress will depend on integrating multimodal approaches and foundational techniques such as culturing to confirm viability. Clarifying the mechanism through which these microbes interact with the tumor microenvironment will be critical to unlocking their therapeutic potential and improving outcomes in TNBC and beyond.