Comparison of vaginal microbiota in gynecologic cancer patients pre- and post-radiation therapy and healthy women

Abstract

Background: While the importance of commensal microbes in vaginal health is well appreciated, little is known about the effects of gynecological cancer (GynCa) and radiation therapy (RT) on the vaginal microbiome (VM) of postmenopausal women.

Methods: We studied women with GynCa, pre- (N = 65) and post-RT (N = 25) and a group of healthy controls (N = 67) by sequencing the V4 region of the 16S rRNA gene from vaginal swabs and compared the diversity and composition of VMs between the three groups accounting for potential confounding factors in multivariate analysis of variance.

Results: Comparisons of cancer vs healthy groups revealed that Lactobacillus and Bifidobacterium have significantly higher relative abundance in the healthy group, while the cancer group was enriched in 16 phylogroups associated with bacterial vaginosis (BV) and inflammation, including Sneathia, Prevotella, Peptoniphilus, Fusobacterium, Anaerococcus, Dialister, Moryella, and Peptostreptococcus. In our sample, RT affected the α-diversity and correlated with higher abundance of typically rare VM species, including several members of the Lacnospiraceae family, a taxon previously linked to vaginal dysbiosis. In addition to cancer and treatment modalities, age and vaginal pH were identified as significant parameters that structure the VM.

Conclusions: This is among the first reports identifying VM changes among postmenopausal women with cancer. RT alone seems to affect several phylogroups (12 bacterial genera), while gynecological cancer and its treatment modalities are associated with even greater significant shifts in the vaginal microbiota including the enrichment of opportunistic bacterial pathogens, which warrants further attention.

Keywords: 16S rRNA gene; gynecologic cancer; postmenopausal women; radiation therapy; vaginal microbiota.

Conflict of interest statement

The authors declare no conflict of interest.

© 2020 The Authors. Cancer Medicine published by John Wiley & Sons Ltd.

Figures

FIGURE 1

Vaginal microbiota in cancer (pre‐RT) and healthy groups. A, Heatmap showing hierarchical clustering of differentially abundant bacterial genera between the healthy and cancer group (n = 20). Three groups were observed: cluster I contained three genera significantly enriched in the healthy group, which included Bifidobacterium, Allistipes, and Lactobacillus. Cluster II grouped eight bacterial genera which were found in both groups but showed higher abundances in the cancer group overall. Finally, cluster III grouped nine genera which were observed mostly enriched in the GynCa group and completely absent from the majority of the healthy samples. B, Boxplots showing the estimated median relative abundance of Lactobacillus spp. in healthy and pre‐RT groups. Note that, in general, Lactobacillus spp. was more abundant in healthy than pre‐RT samples. C, Boxplots showing the estimated median relative abundance of Sneathia spp. in healthy and pre‐RT samples. A higher relative abundance of Sneathia spp. was observed in pre‐RT samples vs healthy. In panels (B) and (C), boxplots represent the first and third quartile and the horizontal segment represent the median value. RT, radiation therapy

FIGURE 2

Comparison of community diversity metrics among healthy, pre‐RT and post‐RT cancer vaginal microbiome communities. A diversity overall increases from healthy to pre‐ and post‐RT cancer groups, a potential indication of disturbance in the microbial communities. Cancer samples show higher diversity than healthy samples in terms if richness, Shannon, and phylogenetic diversity. Post‐RT samples show slightly higher diversity than post‐RT samples in terms of Shannon and phylogenetic diversity. RT, radiation therapy

FIGURE 3

Differentially abundant bacterial operational taxonomic units between (A) healthy and cancer and (B) pre‐ and post‐RT vaginal microbiomes. Previously reported associations for each phylogroup are marked with squared indications and described in the discussion section. RT, radiation therapy

FIGURE 4

Detection of discriminative phylogroups in healthy vs cancer cohort. The plot compares the estimated relative abundance of eight selected operational taxonomic units discriminative of healthy (Lactobacillus, Allistipes, Akkersmania, and Bifidobacter) or GynCa groups (Prevotella, Porhyromonas, Fusobacteria, and Peptococcus). Continuous black lines correspond to the estimated mean relative abundances while striped black lines depict the median