Cell: The first pan-cancerous fungal microbiome map based on 35 cancer types

Recently, a research team led by the Weizmann Institute of Science in Israel comprehensively characterized the fungal communities of 35 types of cancer in tissues and blood, and constructed the first pan-cancer fungal microbiome map
.
The research team compared the fungal community with a matching bacterial and immune group and found bilingual ecosystems where the two coexist, often with permissive rather than competitive microenvironments and unique immune responses
.
Clinical evaluation results indicate prognostic and diagnostic capabilities of tissue and plasma mycobacterial populations, even in stage I cancers, and synergistic predictive performance
with microbiota.
The results of the study were published in Cell in an article titled "Pan-cancer analyses reveal cancer-type-specific fungal ecologies and bacteriome interactions.
"

The article was published in Cell

The research team analyzed fungal DNA from two cohorts of cancer samples that had been tested for bacteria, including FFPE or cryotumor and normal adjacent tissue (NAT) and DNA
extracted from non-cancerous normal breast tissue.
The researchers used four staining methods with different sensitivity and specificity to stain samples of different cancer types (melanoma, pancreatic, breast, lung, and ovarian) to study the presence of fungi (Figure 1).

To account for potential contamination of environmental fungi or the introduction of fungal DNA through sample handling and processing, the research team conducted a negative control analysis to control for fungal contaminants
capable of detecting and removing them.

Figure 1.
Fungal nucleic acids are present in
human cancer.

Staining results showed specific localization patterns
for different cancer types.
In pancreatic, breast and ovarian cancers, fungi are mainly found in cancer cells, while in melanoma and lung cancer, macrophages
are mainly present.

At the same time, tumor bacterial richness is significantly higher than fungal abundance (Figure 2), similar
to the gut microbiome.
In amplicon (WIS) and shotgun metagenomic (TCGA) cohorts, the richness of fungal flora varied significantly
between cancer types.
Interestingly, of the seven cancer types common to these two cohorts, four showed a significant positive correlation between fungal and bacterial abundance within tumors, suggesting that there was no competitive relationship
.
This correlation
was not observed in the negative control.
In addition, a comparison of tumor tissue and its normal adjacent tissue samples showed that there were similar fungal compositions between the two, and there were certain differences
in the fungal composition of samples in different parts.

Figure 2.
Different cancer types exhibit different microbiota
.

Unsupervised analysis revealed three distinct fungal-bacterial immune clusters driven by fungal co-occurrence, named F1 (Malassezia), F2 (Aspergillus-Candida), and F3 (multiple genera including Jersella) (Figure 3).

F1 and F2 contain fewer but more common fungal genera
.
The analysis found that F1, F2, and F3 were associated with different host immune responses, suggesting that different intratumor microbiomes may trigger different host responses
.
In addition, the ratio of certain fungal genera can predict the survival rate
of patients with pan-carcinoma species well.

Figure 3.
The pan-carcinotype
was established by the microbiota-microbiota-immunogroup interaction.

The research team tested whether machine learning (ML) could distinguish cancer types
based on microbiota data.
Evaluating different cancer types on batch-corrected fungal community species revealed that the model could effectively distinguish between 32 cancer types (Figure 4).

At the same time, the fungal DNA analysis results of TCGA blood samples showed that the fungal signal was tumor-specific and synergistic with the bacterial signaling signature, suggesting that the blood fungal group may be used for early cancer detection
.

Figure 4.
Machine learning (ML) analysis reveals cancer type-specific tumor and blood fungal flora
.

Finally, the research team explored the diagnostic and prognostic capabilities
of the cancer microbiota.
Using the WIS cohort, the research team first tested the relationship of
fungi to disease phenotypes, patient survival, and treatment response.
In breast cancer, the team found that Mycobacterium globuli and Mycomyces, previously reported in breast cancer, were enriched in
tumors in patients ≥ 50 years of age.
In lung cancer, the study found that current smokers had higher
intratumoral fungal richness and enrichment of Aspergillus and Mushroom fungi compared to non-smokers.

Research summary diagram

Dr Ravid Straussman, co-corresponding author of the Weizmann Institute for Science, said: "The discovery that fungi are ubiquitous in human tumors should prompt us to better explore their potential implications and re-examine almost all of what we know about
cancer through a 'microbiome lens'.
”

Co-corresponding author Dr Rob Knight of the University of California, San Diego, said: "The discovery of fungi in most human cancers is both surprising and exciting
.
Because we don't know how fungi can get into tumors
.
This is expected because it fits the pattern of a healthy microbiome throughout the body, including the gut, mouth, and skin, where bacteria and fungi interact
as part of a complex community.
”

The study is the first to analyze
the plasma flora of untreated early-stage cancer.
While the source of cell-free plasma-derived fungal or bacterial DNA remains unknown, the microbiome as a whole is a key part of cancer biology and could present significant transformative opportunities, not only in cancer detection, but also in other biotechnology applications related to drug development, cancer evolution, minimal residual lesions, recurrence, and aided diagnosis
.

Meanwhile, another article in the same issue is a collaborative team
from Cornell University Weill Cornell School of Medicine and Duke University.

The researchers also discovered the association between different cancer types and fungal communities and the interaction network between fungi, and identified the corresponding "core bacteria"
.
The results showed that Candida albicans and Saccharomyces cerevisiae are the "core gena" in digestive tract tumors, and they drive multi-species overall changes and interactions
in tumor fungal ecology.
The changes that occur in the tumor fungal ecology affect the tumor immune environment, which in turn promotes the occurrence and development
of tumors.
For example, researchers have found that Candida is associated with upregulation of the expression of pro-inflammatory immune pathways, especially in gastric cancer; In lower gastrointestinal cancers, Candida is associated
with dysregulated gene expression associated with pro-tumor metastasis and cell adhesion.
In this regard, researchers believe that Candida may exert a cancer-promoting effect
by promoting the upregulation of local tumor inflammation, increased permeability of tight junctions, and loss of epithelial barrier function.
In addition, the authors found that the Candida/yeast ratio was lower in early-stage colon cancer and significantly elevated when the disease progressed to stage IV, suggesting that Candida may be strongly associated
with tumor progression.
The results of patient survival analysis showed that the survival rate of patients with high-abundance candida tumors was significantly reduced
.
Finally, the researchers sequenced the primary colorectal samples and isolated and cultured their microorganisms, and after combined RNA detection, it was found that there were live, transcriptionally active candida
bacteria in the tumor tissue of the lower gastrointestinal tract.
Therefore, the authors suggest that therapeutic interventions that target candidal infection and associated inflammation may be potentially effective combination therapies
for cancer.

The two articles show that fungal genes have been found
in tumor tissue, cancer cells, and immune cells within tumors.
But why and how these fungal genes appear in tumor-related tissues is a puzzling question
.
Dr.
Bhatt speculates that perhaps it's because cancer suppresses the immune system, allowing fungi to grow in these tissues/cells while they may be destroyed
by the immune system in other parts of the body.
Or maybe immune cells engulf the fungus and bring the sequence to the tumor site, or even these fungi may originate from microbes in the body/epidermis, and some microbes enter the body tissues
every once in a while.

If fungi do live in tumors, what exactly are they doing? Although these two studies have revealed the correlation between fungi and the occurrence and progression of cancer, the experimental evidence to date has not clearly indicated the specific functions
played by fungi in tumors.
Dr.
Straussman, one of the authors of the study, believes that exploring these questions is not only the key to future research, but also a way to reshape our understanding of cancer – we see cancer not just as a disease, but as an "ecosystem"
.

Resources:

Pan-cancer analyses reveal cancer-type-specific fungal ecologies and bacteriome interactions.
Cell, 2022, doi:10.
1016/j.
cell.
2022.
09.
005.

https://doi.
org/10.
1016/j.
cell.
2022.
09.
005