Introduction American women of African or Hispanic ancestry have increased risk of vaginal microbiome dysbiosis compared to women of European or Asian ancestry. However, the association between vaginal microbiome composition and ethnicity within Europe is largely unknown. We investigated this association in Amsterdam, The Netherlands.

Methods Non-pregnant women (18–34 years, n = 564) representing six ethnic origins (Dutch, South-Asian/Indonesian Surinamese, African Surinamese, Ghanaian, Turkish, and Moroccan) were cross-sectionally selected from the ongoing HELIUS multi-ethnic cohort study in Amsterdam for vaginal microbiome analysis. Extracted DNA from self-sampled vaginal swabs was sequenced targeting the V3V4 region of the 16S rRNA gene and using the Illumina MiSeq platform, and sequence reads were clustered using hierarchical clustering.

Results Clustering of 502/564 samples with sufficient read counts resulted in microbiome clusters dominated by Lactobacillus crispatus (n = 120), L. iners (n = 168), L. jensenii (n = 8), L. gasseri (n = 10), Streptococcus agalactiae (n = 8), Bifidobacteriaceae/Bifidobacterium spp. (n = 10), Gardnerella vaginalis (n = 78), and a mixture of anaerobes (n = 100), respectively. Microbiome composition was significantly associated with ethnic origin (P = 0.002). Women of Dutch ethnic origin had the highest prevalence of L.crispatus-dominated microbiome (40% vs 16–26% in the other ethnic groups), the lowest prevalence of L. iners-dominated microbiome (28% vs 31–39% in the other groups), and the lowest prevalence of clusters dominated by G. vaginalis or a mixture of anaerobes (25% vs 30–45% in other groups). Turkish women and South-Asian/Indonesian Surinamese women had the highest prevalence of L. iners-dominated microbiome (38% and 39%, respectively), and women from African descent (African Surinamese and Ghanaian women) the highest prevalence of clusters dominated by G. vaginalis or a mixture of anaerobes (48% and 44%, respectively).

Conclusion This large multi-ethnic study shows that dysbiotic vaginal microbiome compositions are significantly increased in women of non-Dutch ethnic origin. Therefore, these women may be at increased risk of STI acquisition and adverse reproductive health outcomes.

Disclosure of interest statement The HELIUS study is funded by the Academic Medical Centre Amsterdam, the Public Health Service of Amsterdam, the Dutch Heart Foundation (project number 2010T084), the Netherlands Organisation for Health Research and Development (ZonMw; project number 200500003), and the European Union (FP-7; project number 278901). The vaginal microbiome analyses were funded by the Aids Fonds Netherlands (project number 201102). The authors declare no conflicts of interest.