Abstract
Background
A number of randomized controlled trials have found that doxycycline post exposure prophylaxis (PEP) can reduce the incidence of gonorrhoea, chlamydia and syphilis in men who have sex with men (MSM). If tetracycline resistance is associated with resistance to other antimicrobials in a range of bacterial species, then doxycycline PEP could have the unintended effect of selecting for resistance to other antimicrobials in these bacterial species.
Methods
Antimicrobial susceptibility data were retrieved from two sources: pubMLST (https://pubmlst.org/) and Pathogenwatch (https://pathogen.watch/) for the following bacterial pathogens: Klebsiella pneumoniae, Salmonella enterica subsp. Enterica serovar Typhi, Campylobacter jejuni, Staphylococcus aureus, Streptococcus pneumoniae and Streptococcus pyogenes. We assessed if tetracycline resistance was associated with resistance to six relevant antimicrobials.
Results
We found evidence of cross resistance to various antimicrobials in all six bacterial species assessed. Cross resistance was found in 4 of 5 antimicrobials for K. pneumoniae, 1 of 2 for C. jejuni, 3 of 5 for S. enterica subsp. Enterica serovar Typhi, 5 of 5 for S. aureus, 5 of 6 for S. pneumoniae and 2 of 3 for S. pyogenes. These associations include a higher prevalence of methicillin resistance in tetracycline resistant S. aureus, penicillin resistance in S. pneumoniae, macrolide and clindamycin resistance in S. pyogenes, fluoroquinolone resistance in S. enterica subsp. Enterica serovar Typhi and third-generation cephalosporin resistance in K. pneumoniae.
Conclusion
These results suggest that studies evaluating the effects of doxycycline PEP should include the effects of doxycycline on resistance not only to doxycycline but also to other antimicrobials and in a broader array of bacterial species than has been included in doxycycline PEP studies thus far.
Introduction
Doxycycline, a tetracycline antimicrobial, can be taken after condomless sex to reduce the probability of acquiring a bacterial sexually transmitted infection from this sexual encounter. Four randomized controlled trials have found that this doxycycline post exposure prophylaxis (PEP) can reduce the incidence of gonorrhoea, chlamydia and syphilis in men who have sex with men (MSM).1–3 This has led to a number of centres around the world to offer, or consider offering, doxycycline PEP to a proportion of MSM attending their clinics.4–6
An important concern about the introduction of doxycycline PEP is that it will induce resistance to tetracyclines in various other bacterial species. Two doxycycline PEP studies have evaluated the effect of doxycycline on tetracycline resistance in Neisseria gonorrhoeae. Both found no statistically significant effect, but the duration of follow-up was short, and the number of gonococcal isolates tested were small (n=9 1 and n=47 3 ). Interim data presented at CROI 2023 7 reported an increase in the prevalence of doxycycline resistance in Staphylococcus aureus in the DoxyPEP arm and an increased prevalence of doxycycline resistance in commensal Neisseria species compared to the control group at 12 months. 8
In addition to selecting for resistance to tetracyclines, doxycycline PEP could select for resistance to other classes of antimicrobials.9–11 This effect could be direct or indirect. Tetracyclines have been noted to
In this manuscript, we extend the cross-resistance to other important bacterial pathogens: ESKAPE pathogens, Shigella spp., Salmonella spp., Campylobacter spp., Streptococcus pneumoniae and S. pyogenes.
Materials and methods
Bacterial collections
Descriptive information about the datasets used and their provenance.
NA-not available;
‘N isolates’ refers to the number of isolates in the collection with tetracycline susceptibility results.
P= phenotypi, G genotypic. For further study details please see STable 7.
Collections were included if they had antimicrobial susceptibility data that included tetracyclines for at least 400 isolates (Figure 1). Pathogenwatch permits the download of data for up to 5000 isolates per species. PubMLST does not have a limit. We downloaded the most recent data of up to 5000 isolates for each species until April 2023. The three species with available data from Pathogenwatch that met our entry criteria were: Salmonella enterica subsp. Enterica serovar Typhi, S. pneumoniae and S. aureus. In the case of K. pneumoniae, no antimicrobial susceptibility data was available in this download-for-the-most-recent-5000-isolates option on Pathogenwatch, and we thus used the single collection of this species that is available on Pathogenwatch, where the susceptibility data is available. This EuSCAPE collection consists of 1505 isolates of K. pneumoniae from patients in 244 hospitals in 32 countries during the European Survey of Carbapenemase-Producing Enterobacteriaceae between November 2013 and April 2014.
17
The pubMLST dataset had relevant data for Campylobacter jejuni, S. pyogenes and S. pneumoniae. The pubMLST data for S. pneumoniae only included susceptibility data for tetracyclines and penicillin. Since Pathogenwatch did not contain penicillin susceptibility data we used both the Pathogenwatch and pubMLST data for S. pneumoniae. Flowchart illustrating study methodology.
For both collections, we used dichotomized resistant/susceptible variables for antimicrobial susceptibility. For the pubMLST collections of S. pneumoniae, S. pyogenes, and C. jejuni the susceptibility data included an intermediate category for a small minority of isolates (typically <3% of isolates). These were reclassified as resistant in our analyses. Omitting these results made little difference to the results (data not shown).
We analyzed data for the following antimicrobial classes whenever these were reported in the datasets used: penicillin, ampicillin, methicillin, aminoglycosides/kanamycin, fluoroquinolones/ciprofloxacin, third-generation cephalosporins (3Gcephalosporins), macrolides/azithromycin/erythromycin, sulphonamides and clindamycin. These antimicrobials were selected based on their clinical importance and risk of AMR emergence.
Data analysis
Statistical analyses were conducted using Stata v16.1 (StataCorp, LLC College Station, Texas). The Chi-squared test was used to compare groups.
Results
Antimicrobial susceptibility data was available for six bacterial species (Table 1). We found evidence of cross resistance to various antimicrobials in all six bacterial species that were assessed. Cross resistance was found in 4 of 5 antimicrobials for K. pneumoniae, 1 of 2 for C. jejuni, 3 of 5 for S. enterica, 5 of 5 for S. aureus, 5 of 6 for S. pneumoniae and 2 of 3 for S. pyogenes.
K. pneumoniae
Prevalence of resistance to various classes of antimicrobials by presence or absence of tetracycline resistance in six pathogenic bacterial species (column percentages).
S- susceptible; R- Resistant; Column percentages;*< 0.01 **p < .001; NA – Not Applicable; # data for penicillin susceptibilities is from pubMLST, all other data is from Pathogenwatch.
S. enterica subsp. enterica serovar Typhi
The prevalence of tetracycline resistance was low (161/4949; 3.3%). Resistance to three of five antimicrobial classes assessed was more prevalent in tetracycline resistant isolates than susceptible isolates: ampicillin (95.0%/23.8%), ciprofloxacin (98.1%/89.1%), and sulphonamides (99.4%/24.5%), all p-values <0.001. Resistance to 3Gcephalosporins was less prevalent in tetracycline resistant isolates (1.2%/12.4%; p-value <.001).
Campylobacter jejuni
Tetracycline resistance was present in 602/1396 (43.1%). Resistance to fluoroquinolones (but not macrolides) was more prevalent in tetracycline resistant isolates than susceptible isolates: (81.6%/19.8%, respectively; p-value <.001).
S. aureus
Tetracycline resistance was evident in 330/3232 (10.2%). Resistance to five of five antimicrobials assessed was more prevalent in tetracycline resistant than susceptible isolates: penicillin (97.3%/78.0%), methicillin (81.2%/33.4%), ciprofloxacin (39.1%/21.4%), amikacin (50.0%/26.2%) and erythromycin (65.2%/36.3%), all p-values <0.001).
S. pyogenes
Three quarters of isolates (230/301, 76.4%) were resistant to tetracyclines. These isolates had a higher prevalence of resistance to erythromycin and clindamycin (but not penicillin) than tetracycline susceptible isolates (39.6%/14.5% and 33.6%/8.2%, respectively, all p-values <0.001).
S. pneumoniae
The prevalence of tetracycline resistance was 568/1389 (59.1%) in the pubMLST collection and 415/3019 (12.1%) in Pathogenwatch. Resistance to five out of six antimicrobials was more prevalent in tetracycline resistant isolates than susceptible isolates: penicillin (45.9%/1.5%), kanamycin (1.2%/0.0%), sulphonamides (49.4%/14.0%), clindamycin (74.5%/0.7%) and erythromycin (87.2%/19.2%), all p-values <0.001).
Discussion
We found evidence of cross resistance to 9 out of 9 antimicrobials in all six bacterial species assessed. In K. pneumoniae, cross resistance was found in 4 of 5 antimicrobials assessed, in C. jejuni in 1 of 2, in S. enterica subsp. Enterica serovar Typhi 3 of 5, in S. aureus 5 of 5, in S. pneumoniae 5 of 6 and in S. pyogenes 2 of 3 antimicrobials. Only one example of a negative association between tetracycline resistance and another class of antimicrobials was found – 3Gcephalosporins in S. enterica subsp. enterica serovar Typhi.
A number of the positive associations are of particular clinical relevance. These include the higher prevalence of methicillin resistance in tetracycline resistant S. aureus, penicillin in S. pneumoniae, macrolide and clindamycin resistance in S. pyogenes, fluoroquinolone resistance in S. enterica subsp. enterica serovar Typhi and 3Gcephalosporin resistance in K. pneumoniae.
These positive associations raise the concern that doxycycline PEP could select for resistance to these other classes of antimicrobials in a range of important bacterial pathogens. It is important to note that the only doxycycline PEP study to evaluate the effect of doxycycline on the prevalence of MRSA did not find that doxycycline resulted in an increase in MRSA. 8 A number of caveats are important in interpreting this PEP study. The follow-up was relatively short, the numbers in both arms were relatively small, and there was ample opportunity for transmission of S. aureus between participants in the study arms. 8 In the four doxycycline PEP RCTs, the average doxycycline consumption was 4 to 16 doses of 200mg doxycycline per month in the PEP arms.1–3 This consumption is 110- to 440-fold higher than the mean consumption of tetracyclines in a typical HIV PrEP cohort. 18 These high levels of consumption mean it will be important to follow up study participants for a longer time before one can conclude that there is no risk of bystander selection.
There are important limitations to our study. The sample collections are not representative collections of circulating strains of each species. Rather they are heterogenous samples that have been collected from single or multiple countries, for short or longer periods, from only hospitalized patients (K. pneumoniae) or mixed groups of patients. These collections may be more likely to contain isolates with AMR than representative samples from the general population. We do not have details of antimicrobial exposure of the individuals from which the isolates were obtained or the methodology used to assess antimicrobial susceptibility. In some of the datasets we used, there were slight differences in the number of isolates that had their susceptibility assessed for the different antimicrobials. For example, if we consider the C. jejuni collection from pubMLST, for example, a total of 1366 isolates had the combination of tetracycline and fluoroquinolone susceptibility assessed, whereas 1542 isolates had the combination of tetracycline and macrolides assessed. These differences would, however, typically result in a misclassification bias which typically reduces the strength of associations between variables. 19 It is also important to note that the use of doxycycline as PEP could, by reducing the incidence of various STIs, reduce the use of more broad-spectrum antimicrobials such as ceftriaxone and azithromycin 3 . This may more than offset the selection pressure produced by doxycycline PEP. The effect of doxycycline may vary according to the degree of tetracycline resistance and specific genetic background and genetic mutation conferring tetracycline resistance. Finally, tetracycline resistance does not equate with doxycycline resistance. Whilst a number of studies have found a strong positive correlation between MICs for tetracycline, minocycline and doxycycline for a number of bacterial species this is not necessarily the case for all species.20,21 We therefore need to exert due caution when extrapolating from tetracycline cross resistance to doxycycline cross resistance.
Furthermore, we only evaluated cross-resistance in a limited number of species. A recent surveillance study of XDR isolates of S. sonnei from France found that 87% were resistant to tetracyclines. 15 This and other outbreaks of multi resistant S. sonnei have been linked to sexual transmission within international sexual networks of men who have sex with men.15,22 The authors of the French study postulated that the introduction of doxycycline PEP may provide an additional selection pressure for XDR strains of S. sonnei in this population. 15 We would agree and go further to suggest that studies evaluating the effects of doxycycline PEP should include the effects of doxycycline on resistance not only to doxycycline but also to other antimicrobials and in a broader array of bacterial species than has been included in doxycycline PEP studies thus far.
Supplemental Material
Supplemental Material - Doxycycline post exposure prophylaxis could select for cross-resistance to other antimicrobials in various pathogens: An in silico analysis
Supplemental Material for Doxycycline post exposure prophylaxis could select for cross-resistance to other antimicrobials in various pathogens: An in silico analysis by Zina Gestels, Sheeba Santhini Manoharan-Basil and Chris Kenyon in International Journal of STD & AIDS
Footnotes
Acknowledgements
We would like to thank the scientists who shared their data on Pathogenwatch and PuBMLST.
Authors’ contributions
CK, ZG and SMB conceptualized the study. CK was responsible for the statistical analyses and writing the first draft. All authors read and approved the final draft.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
