Background:
Doxycycline post-exposure prophylaxis (doxy-PEP) reduces bacterial sexually transmitted infections in men who have sex with men (MSM) and transwomen (TW) living with HIV or on HIV pre-exposure prophylaxis (PrEP). The impact on the gut bacterial microbiome and its associated antimicrobial resistance genes (ARGs) is unknown.
Methods:
The DoxyPEP trial compared doxy-PEP use (DP) to standard of care (SOC) for 501 MSM/TW living with HIV or on PrEP. Metagenomic sequencing of DNA and RNA (DNA-seq, RNA-seq) was performed on self-collected rectal swabs from enrollment (M0) and month 6 (M6) from 50 SOC and 100 DP participants matched by HIV infection status. We analyzed all samples passing minimum sequencing quality control standards. From DNA-seq data, we compared bacterial microbiome diversity and total bacterial abundance in 59 DP and 30 SOC participants, and assessed baseline tetracycline (TCN) ARG abundance at enrollment. We then compared changes in ARG expression at M0 and M6 using RNA-seq data of samples from 46 DP and 24 SOC participants. Bacterial abundance was normalized per million reads sequenced (reads per million, rpm); ARG abundance and expression were additionally normalized by gene length (depth per million, dpm). P-values were obtained by the Wilcoxon rank-sum test and adjusted for multiple comparisons. Linear regression tested the association between number of DP doses taken since enrollment to TCN ARG expression.
Results:
Gut bacterial microbiome alpha diversity, beta diversity, and total bacterial abundance did not differ between SOC and DP arms at M0 or M6, or over time by arm. TCN ARGs were detected in all M0 DNA-seq samples, with 46% of all ARG abundance derived from TCN ARGs. Median TCN ARG expression increased 4.4-fold between M0 and M6 in the DP group (0.07 to 0.31 dpm; p<0.01) and was 5-fold higher at M6 in the DP versus SOC group (0.31 versus 0.06 dpm; p<0.01; Figure). There were no significant differences in ARG expression between M0 and M6 DP samples for non-TCN class ARGs. DP participants with M6 RNA-seq samples took a median of 42 DP doses between M0 and M6 (IQR: 29-65 doses), and we found that a higher number of DP doses was associated with greater TCN ARG expression (p<0.01).
Conclusions:
Doxy-PEP did not significantly alter bacterial gut microbiome diversity or composition but was associated with an increase in TCN ARG expression, without affecting non-TCN ARG classes. The clinical impact of these findings remains undetermined and should be further studied.
