This paper is part of NMJ's 2018 Microbiome Special Issue. Download the full issue here.
Hieken TJ, Chen J, Hoskin TL, et al. The microbiome of aseptically collected human breast tissue in benign and malignant disease. Scientific Reports. 2016;6:30751.
To determine resident microbiome differences in breast tissue vs skin and in malignant vs nonmalignant breast tissue samples.
Observational cohort study
Thirty-three women scheduled to undergo breast surgery at Mayo Clinic had their postsurgical specimens analyzed. Roughly half of the women were found to have breast cancer (n=17), and half were diagnosed with benign breast disease (BBB; n=16). All of those with breast cancer were estrogen- and progesterone-receptor–positive, and 29% were HER2/neu receptor–positive (n=4). One participant with cancer dropped out of the analysis. Of the 15 participants with breast cancer, 10 had stage I and 5 had stage II disease, and 13% of all of those with breast cancer had lymph node involvement.
Notably, there were some differences in the characteristics of the 2 groups (women with cancer and women with BBB). First, the median age of each group and, correspondingly, menopausal status, was significantly different. The overall median age of the cohort was 60 (range, 33-84); the median age was 75 (range, 44-84) for women with invasive cancer vs 49 (range, 33-70) for women with BBD (P=0.001). Of the women with cancer, 86.7% were peri/postmenopausal and 13.3% were premenopausal, while 53.9% of the women with BBD were peri/postmenopausal and 46.2% were premenopausal (P=0.02). The time from incision to sample collection was also statistically different between the 2 groups (median 82 min vs 52 min in those with cancer and those without, respectively; P=0.0001).
The presence of a shift in flora before the disease is present means that someday we may be able to stratify risk of developing breast cancer based on the microbiome present in the tissue.
Study parameters assessed: Intraoperative tissue samples of the breast and overlying skin were analyzed using 16S rDNA tag sequencing for microbial DNA signatures. Buccal swabs and breast skin swabs were also obtained and analyzed in the same manner.
Distinct microbial communities existed in the breast tissue vs samples of overlying skin tissues, breast skin swabs, or buccal swabs. When comparing women with cancer to those with BBD, distinct microbial community differences were found. Specifically, several taxa that are less abundant overall are enriched in the cancer tissue vs the BBD tissue, including Fusobacterium, Atopobium, Gluconacetobacter, Hydrogenophaga, and Lactobacillus. Lastly, the nearby disease-free tissue in those with cancer versus the nearby normal tissue in those with BBD differed in taxa significantly (P=0.009).
The authors’ first assertion is that this study “confirms the existence of a distinct breast microbiome and differences between the breast tissue microbiome in benign and malignant disease.” The first part of this may be little news to natural medicine practitioners, who have been affecting the health of nursing babies by modifying mom’s flora, or recommending that a small dusting of infant probiotic be placed on the nipple before feedings. We’ve assumed organisms come from the breast for a long time. Perhaps we’ve based this knowledge on the 2 studies from the 1980s1,2 that suggested the existence of a distinct breast flora, or perhaps we just believed in the absence of evidence. According to the current study’s authors, the 1980s studies that found distinct bacteria inhabiting the breast were widely dismissed, with detractors suggesting the bacteria were likely contaminants from the skin.
Interestingly, while the existence of endogenous bacteria in the breast appears to be news in medicine, it also seems to have been an “open secret” in plastic surgery circles. These bacteria have been suspected to be the cause of a subclinical infection responsible for post-implant capsular contracture.3 Regardless, the study reviewed here confirms our long-held assumption that the breast has its own unique microbiome. That much is crystal clear.
The more intriguing aspect of the study reviewed here is the presence of distinct microbes in cancerous breast tissue vs BBB. The dominant taxonomy was not different, Bacteroidetes and Firmicutes dominated both samples. The differences were in the higher levels of normally very low-abundance flora: Fusobacterium, Atopobium, Hydrogenophaga, Gluconacetobacter, and Lactobacillus (P<0.05). The last one may catch our attention, given Lactobacillus spp are assumed to be beneficial organisms. Lactobacilli, like all of these bacteria, are only associated with the cancer, not causative. The function of these bacteria and precisely how they are interacting with the various components of the stroma is not known, yet.
There were 2 prior studies using molecular (rather than culture) techniques to analyze breast cancer tissue. Xuan and colleagues looked at breast cancer tissue versus normal tissue from the same donor and found that Methylobacterium radiotolerans was enriched in cancerous tissue while Spingomonas yanoikuyae was enriched in the normal controls.4 Further, they found that the diversity of the flora was inversely associated with the extent of disease, with advanced-disease patients having less diversity in the breast biome. This was a very small study, however, with only 20 participants, and it was critiqued by the authors of the study reviewed here as having high potential for contamination due to methodological reasons.
The second study, published by Urbaniak and colleagues, looked at the breast microbiome in 81 women from Canada and Ireland, with and without breast cancer.5 The study was designed to definitively determine whether there were live bacteria (not just their DNA) present in what was has been presumed to be sterile breast tissue. The group did find bacteria, both through molecular and culture techniques, with Proteobacteria the dominant phylum. As an aside, this is also the dominant phylum found in human breast milk.6 The study was not designed to assess differences between normal and cancerous tissue, nor between Canadian and Irish women.
Since publication of the study reviewed here, Wang and colleagues have confirmed that the breast microbiome in women with breast cancer is distinct from the microbiome in normal breast tissue. The microbiome of the mouth, urinary tract, and breast tissue was determined in 57 women with cancer and 21 women without cancer.7 The authors found that the breast microbiome was significantly different between the 2 groups (P=0.03), driven primarily by the presence of Methylobacterium in the cancerous tissue. In addition, several gram-positive organisms including Corynebacterium (P<0.01), Staphylococcus (P=0.02), Actinomyces (P<0.01), and Propionibacteriaceae (P<0.01) were more abundant. Unlike the current study, Lactobacillus spp were not enriched in the breast cancer tissue. However, the presence of Lactobacillus in the urine of postmenopausal women was lower than that of premenopausal women. Oral microbiomes did not differ.
Note that while there are some consistencies in the above molecular studies, much of the data thus far is not consistent. This is due to several factors, including the immense complexity of the microbiome, inherent differences in techniques, expected ethnic variations in biomes, and the low number of participants in each study. Put together, we can confidently say that there is a unique microbial niche in the breast itself, and breast cancer is distinctly different in its microbiome signature vs normal breast tissue. The details of these 2 findings will continue to be flushed out going forward.
A unique aspect of the study reviewed here is that the non-diseased tissue near the malignancy also harbored a distinct flora when compared with nearby tissue in those with BBB. This is intriguing. The presence of a shift in flora before the disease is present means that someday we may be able to stratify risk of developing breast cancer based on the microbiome present in the tissue. This would be a means of better determining risk of sporadic breast cancer.
In keeping with the popular metaphor of the body’s microbial niches as ecosystems, integrative practitioners are uniquely trained to improve the breast flora in the context of overall health. In the modern reductionist medicine model, singular strains will be touted as specific for breast health. Indeed, there are numerous patented therapeutic probiotics available following this line of thought.8 This would be the equivalent of spreading a single plant seed, or a mere handful of plants, and expecting a complex and healthy ecosystem to arise. While certain strains may eventually emerge in the research, it will always be the entire environment of the body that must be tended for proper establishment of the microenvironment of the breast and its microbiota.
This is not to say that application of particular bacterial strains is never indicated. Several strains of Lactobacillus have been associated with increased immune recognition, decreased tumor growth, and increased survival in rodent models of breast cancer.9 Among these are specific strains of L casei, L plantarum, and L reuteri. This is interesting given that Lactobacillus spp were enriched only in the tissue with breast cancer in the current study. The role of the bacteria, again, has yet to be determined.
For now, there is no outcome data in humans to suggest there are specific probiotics that will help prevent breast cancer or its recurrence. In the absence of evidence, we often fall back to our philosophically based understanding of health and disease. In brief, this can be understood as optimizing the overall health of the organism, providing all necessary components of elements interwoven into the larger landscape of life on the planet. In the context of the breast microbiome, this is certainly our best bet.