Cancer and the Microbiome

As our understanding of the microbiome’s impact on human health continues to grow, its relationship within the context of varied disease pathophysiologies and treatment response is becoming clearer. Though early studies of the microbiome have primarily focused on diseases of the gut, such as inflammatory bowel disease, attention is increasingly being placed on understanding potential links between the microbiome and a wider range of disease states, including cancer. Despite a growth in publications investigating the microbiome in general – and links between the microbiome and oncology in particular – the rate of high-impact publications has lagged behind the broader field of microbiome research.  Nevertheless, such studies have yielded noteworthy results that may profoundly impact the prevention, treatment and prognostication of cancer.

Building the case

Researchers are beginning to elucidate how microbial composition impacts whether tumors develop, how they progress, and whether a given therapeutic is effective. In one study, investigators looked into how diet affects the microbiome of the mammary glands and the subsequent development of breast cancer. Female non-human primates fed a diet that mimics key aspects of the Mediterranean diet (characterized by protein and fats derived mainly from plant sources and relatively high mono-unsaturated fatty acids) exhibited a much higher abundance of Lactobacillus in their breast tissue than those fed a Western diet (high in saturated fats and sodium). A higher Lactobacillus volume was shown to have a negative correlation with malignant tumors (1). Such findings may implicate the mammary gland microbiome as a potential avenue for breast cancer treatment.

The microbiome has already been shown to affect patient response to immune checkpoint inhibitor (ICI) therapy – a treatment hailed for extending survival of some cancer patients by years; in some cases, even eliminating all signs of disease in some advanced clinical cancer cases. The ICI approach, however, has proven ineffective for many cancer patients with different tissue pathologies. As a result, several investigations now underway aim to better understand the mechanisms by which gut microbiota impact ICI efficacy. In one such study, a team in China conducted a retrospective review of data on 109 patients with non-small cell lung cancer (NSCLC) undergoing treatment with an anti-PD-1 (Programmed cell death protein 1), a subset of whom had been administered antibiotics concurrently. Oral treatment with antibiotics (well-documented as having an effect on gut microbial composition) was associated with significantly lower progression-free survival rates and lower overall survival rates for NSCLC patients on an anti-PD-1 therapy (2).

Several studies published in Science in early 2018 have demonstrated similar results, finding a correlation between ICI efficacy and the presence of certain microorganisms in the gastro-intestinal tract. In a study of patients with advanced epithelial cancers, Routy et al. demonstrated that fecal microbiota transplantation (FMT) from patients who responded well to anti-PD-1 therapy into mice that were either of germ-free health status or treated with antibiotics enhanced ICI treatment efficacy. Conversely, FMT from non-responders did not have the same effect. Specifically, patient response to the anti-PD-1 treatment correlated to an abundance of Akkermansia muciniphila, which is associated with a host of health benefits, including improved metabolic health and reduced incidence of type 2 diabetes. When the mice received oral supplements of Akkermansia muciniphila after FMT from non-responder patients, anti-PD-1 efficacy improved (3).

About 70 percent of the human microbiota is present in the gut. Consequently, much emphasis has been placed on uncovering the link between the microbiome and gastro-intestinal malignancies, such as colorectal cancer (CRC). To gain insights into the underlying mechanisms at work, a group of investigators identified that the metastasis-related secretory protein cathepsin K (CTSK) mediates the imbalance between gut microbiota and CRC metastasis. They then implanted immortalized murine MC38 colon adenocarcinoma cells into the cecal mesentery of healthy wild type mice that were treated with antibiotics, and administered E. coli to promote gut microbiota imbalance. The mesentery of the intestine contains blood vessels, lymphatics and nerves to supply and support intestinal function, and orthotopic injection of these MC38 colon adenocarcinoma cells into the mesentery aimed to replicate a metastatic model. Larger tumors and more liver metastatic foci were found in the mice administered with E. coli, compared with the control group, concomitant with greater CTSK overexpression (4). Additionally, it has proven possible to reduce the spread of colorectal cancer cells by silencing expression or inhibiting activity of CTSK in this animal model.

As our understanding of the immune system’s impact on cancer has improved, investigators have begun to explore how the symbiotic relationship between the microbiome and the immune system may impact anti-tumor immunity. Most recently, Nature Communications reported the role of the gut microbiome in establishing melanoma anti-tumor immunity in RNFD-/- mice). Researchers determined that a cocktail of 11 bacterial strains established anti-tumor immunity and limited the proliferation of melanoma cells (5).

An expanding toolkit

Evidence that the microbiome plays a role in cancer development, progression, and response to therapy continues to mount. Increasingly sophisticated tools will be required to allow researchers to adequately explore this complex relationship and to enable control of potential confounding factors, which is crucial in improving study reproducibility and reducing variability.

To date, the germ-free mouse has served as a valuable tool for pre-clinical research. Ablation of the mouse microbiota, which results in a blank slate that enables the study of host-microbiome interactions, provides a suitable platform for drug candidate screening. Such a model, however, cannot replicate the immune system’s influence on cancer development, or how patients will respond to therapy.

A more expansive toolkit is required – one that goes beyond the reductionist approach of a germ-free model and includes additional resources that support a more holistic methodology. An exciting development in this endeavor is the wild mouse gut microbiome, a diverse microbial composition that has evolved in a natural environment alongside its host. Studies by researchers at the National Institute of Diabetes and Digestive and Kidney Disease have demonstrated that the wild mouse microbiome is significantly different from that of a lab mouse. Most notably, it presents greater resistance to tumor development. Such a model presents numerous opportunities: i) to serve as a useful tool for the investigation and discovery of new methods of protecting against cancer, ii) to determine how a host responds to disease and therapy naturally, and iii) to gain a greater understanding of host-microbe interactions in cancer patients.

The use of such models, incorporating a defined microbiome, is part of a broader shift towards more customizable constructs. A germ-free mouse model affords flexibility, allowing specific tailoring of the host microbiome to research needs and the wider study objective. Starting with a germ-free mouse model that is either humanized (with the reconstitution of human immune cells or through replacement of murine genes with human orthologs) or in which a gene of interest is knocked out, the mouse can be colonized with the microbiota of choice. Then, a study-sized cohort can be bred, using rigid husbandry practices and isolators to maintain the model’s microbial composition. 

Custom microbiota association, typically administered using a FMT from a patient donor to a germ-free model, is also gaining attraction as a personalized medicine strategy. The resulting “humanized” model facilitates investigation into how patient-specific microbiota may impact a variety of biological systems, and how response to oncology therapeutics may differ across patient populations. The use of custom microbiota associations could refine the catalog of the differentially abundant microbes in cancer patients and facilitate mechanistic studies to address causation in translational research and preclinical trials.  

Germ-free mice associated with human microbiota allow researchers to recapitulate a large part of the human gut microbiota composition (100% of phyla, 11/12 classes and ~88% of genus-level taxa) (6), which makes them powerful tools for improving research predictability and translatability.

Yet, even as more sophisticated pre-clinical research tools emerge, challenges remain. The complexity of the tumor microenvironment further complicates treatment; differences in human and murine biology remain a confounding factor in pre-clinical research. Though humanized models are bridging the gap, complete harmonization across the study of microbiota, animal models, and patients is not yet reality. Despite the obstacles, oncologists continue to make great strides in understanding the relationship between the microorganisms we harbor and how cancer develops and responds to treatment. As precision medicine advances, exploring how a patient’s unique microbial composition may impact cancer progression and treatment will likely take a more central stage. And, as more sophisticated mouse models emerge, researchers should be able to better understand how modulation of the microbiome and improved gut homeostasis may improve therapeutic efficacy. Such developments bode well for the future of oncology drug discovery.