Most biomarkers that clinicians use to diagnose cancer are human. But growing evidence for a link between the microbiome and malignancy presents an opportunity to look at the presence of microbial DNA as a way to identify and possibly predict disease. In a study published June 23 in Genome Biology, researchers demonstrate that it is possible to detect bacterial DNA in blood samples in both healthy people and melanoma patients.
“This is yet another manuscript that demonstrates that you can identify microbial signatures in the blood, but there are a lot of nuances with these [identification] strategies to date—particularly when it comes to possible contamination,” says Jennifer Wargo, a melanoma surgeon at MD Anderson Cancer Center in Texas who was not involved in the study. “There’s definitely a lot of promise, but it’s early days and these strategies need to be optimized and, ideally, standardized.”
Stephen Wong, a postdoc in Sarah-Jane Dawson’s group at the Peter MacCallum Cancer Centre in Australia, and colleagues have focused in the past on cancer biomarkers, mainly circulating tumor DNA. After reading previous reports of circulating microbial DNA and its involvement in diseases such as HIV and cancer, the team set out to see whether or not they could find it in human blood samples themselves.
In the new study, the researchers performed 16S sequencing, which catalogs bacterial ribosomal RNA genes, on plasma, stool, and saliva samples from 16 patients with advanced melanoma. Wong says they were initially amazed when they detected a large bacterial signature in plasma, but quickly realized that many of the microbes they detected were common contaminants of the lab environment and reagents.
The team began to run controls without samples included alongside their samples at every step of the process. As part of their bioinformatics analysis, they filtered out the contamination and found a small but seemingly genuine bacterial DNA signal in the plasma samples that was distinct from stool and saliva samples. When the researchers compared cell-free microbial DNA found in 15 healthy people and 15 new melanoma patients, they observed that healthy plasma contained a wider diversity of bacterial species. They also identified one genus—Castellaniella—that was unique to healthy plasma, which indicates that it might someday be possible to distinguish healthy and sick people via circulating microbial DNA.
“The report is yet one more example that we can find microbial signals—in this specific case, actually, bacterial signals—pretty much in every place we look. The challenge is finding out what their biological relevance is,” says Nadim Ajami, who works with Wargo at MD Anderson and was not involved in the study.
16s sequencing only gives information about bacteria and some archaea, but “not forgetting about the other components of the microbial world is also key,” he says. For instance, “the presence of phage can change the biological profile of a bacterium, as phage carry genes that impact bacterial function.” Plus, finding bacterial chromosomes and phage can create a more complete picture of what microbes are present.
“Usually people think about microbes as contamination and not as information,” says Sandrine Miller-Montgomery, who was part of a team that showed in 2020 that microbiome information from blood and tissue is associated with various types of cancer. Miller-Montgomery is now the CEO and president of Micronoma, a biotech company that sprang from the 2020 work and focuses on predicting cancer by analyzing microbial signals in liquid biopsies such as blood samples. “I’m hoping that others are going to look into [the] circulating microbiome for other diseases,” she adds. While “it’s true that it’s low biomass, low biomass doesn’t mean that it’s not super relevant.”