What this study adds: This opinion article discusses doxycycline post-exposure prophylaxis (PEP) in the context of the South African healthcare system. Benefits and concerns of doxycycline PEP are discussed, and recommendation for implementation in controlled settings is provided.

South Africa has one of the largest burdens of bacterial sexually transmitted infections (STIs) worldwide. Estimates of the World Health Organization (WHO) Spectrum STI model are that each year 3.9 million new cases of Chlamydia trachomatis, 2.2m new cases of Neisseria gonorrhoeae and 47 500 new cases of active syphilis infection occur in men in South Africa.¹ In women, the estimated number of annual new cases is 1.9m of C. trachomatis, 2.3m of N. gonorrhoeae, and 23 175 of active syphilis.¹ Men who have sex with men (MSM) are disproportionally affected. Prevalence estimates for C. trachomatis (12% – 26%), N. gonorrhoeae (8.1% – 28%) and active syphilis (5.2% – 17%) reported in research studies of MSM exceed that of the general male population (6.0%, 3.5% and 0.97%, respectively) (Table 1); Mycoplasma genitalium prevalence of 8.1% was reported in MSM.^1,2,3,4,5,6 Incidence data are limited but one study reported a combined incidence of C. trachomatis and N. gonorrhoeae of 188/100 person-years in MSM in Johannesburg.⁵

Interventions are urgently needed to reduce the burden of STIs, associated morbidity and potential impact on HIV transmission. Recently, several randomised controlled trials (RCT) have reported on doxycycline prophylaxis as an effective, well-tolerated, and acceptable novel biomedical approach to prevent bacterial STIs among MSM. We discuss the effectiveness, concerns about antimicrobial resistance (AMR), and considerations for implementation of doxycycline prophylaxis for bacterial STIs in South Africa.

Doxycycline is a bacteriostatic antibiotic, a second-generation tetracycline with broad antibacterial spectrum that acts on the ribosomal protein synthesis unit.⁷ Its liposolubility ensures high bioavailability and high tissue and fluid penetration. For STI post-exposure prophylaxis (PEP), a single dose of 200 mg doxycycline was used, ideally to be taken within 24 h (maximum 72 h) after condomless sex.^8,9

Doxycycline (100 mg twice daily) is used for syphilis treatment of non-pregnant adults in the South African syndromic regimen for genital ulcer disease due to global benzathine benzylpenicillin shortage.¹⁰

Although azithromycin covers C. trachomatis and M. genitalium in the syndromic regimen for genital discharge,¹⁰ in case of diagnostic test, the Southern African HIV Clinicians Society’s guidelines recommend doxycycline for first-line pathogen-directed treatment of C. trachomatis, and as part of the sequential therapy regimen for documented symptomatic M. genitalium infection.¹¹ Doxycycline is not recommended for N. gonorrhoeae due to the high levels of antimicrobial resistance (AMR), with up to 90% of strains being resistant to tetracyclines.^12,13,14

Other than STIs, doxycycline is used at the primary healthcare level for treatment of acne vulgaris, tick bite fever, in patients with an acute exacerbation of chronic obstructive pulmonary disease who have severe penicillin allergy, and as malaria prophylaxis.¹⁵

Three RCTs have demonstrated the efficacy of doxycycline PEP for STI prevention in MSM and one in women (Table 2).^16,17,18,19 The IPERGAY trial from France reported a 47% reduction in STI incidence in the doxycycline PEP group of MSM taking HIV pre-exposure prophylaxis (PrEP).¹⁶ The relative reduction was almost 70% for C. trachomatis and syphilis, but there was no reduction in N. gonorrhoeae. The DOXYVAC trial by the same investigators observed similar reductions in STI in MSM taking HIV PrEP and with a history of STI treatment with an 84% reduction in C. trachomatis and syphilis and 51% reduction in N. gonorrhoeae.¹⁷

The DoxyPEP trial in the United States (US) of MSM and transgender women (TGW) with a history of STIs in the prior year also showed reductions in STI incidence among those using HIV PrEP (66% overall) and in people living with HIV (PLHIV) (62% overall).¹⁸ Participants received three bottles of 30 doxycycline tablets and were counselled to take 200 mg of doxycycline, ideally within 24 h but no later than 72 h after condomless anogenital, vaginal, or oral sex, and not more than one dose every 24 h. In persons taking HIV PrEP, doxycycline PEP reduced syphilis by 87% and chlamydia by 88% while reductions of 77% in syphilis and 74% in chlamydia were observed for PLHIV. Unlike the IPERGAY trial, this study showed a reduction of N. gonorrhoeae incidence of 55% in PrEP users, and 57% among PLHIV. Some of the variability in the efficacy of doxycycline PEP against N. gonorrhoeae in the two French trials compared to the DoxyPEP trial in the US could be explained by the lower background level of doxycycline resistance in the US compared to France at the time of the trials, and that DoxyPEP trial participants were permitted to have more doses than in IPERGAY.²⁰

Unlike the efficacy observed in the three trials among MSM, the dPEP Kenya trial did not show efficacy of doxycycline PEP in cisgender women (18–30 years).¹⁹ A nonsignificant decrease in C. trachomatis was observed, but the number of N. gonorrhoeae cases increased in the doxycycline PEP arm. Furthermore, only one case of syphilis occurred during the study period, making it impossible to draw conclusions on the effectiveness against syphilis. Various explanations for the lack of efficacy in these young women are under investigation; initial results indicate adherence is a major factor with approximately half of 50 randomly selected participants having a hair sample with detectable doxycycline.²¹

The main concern about doxycycline PEP is the effect it may have on AMR, other bacterial pathogens, and on the human microbiome.²⁰ The IPERGAY and DoxyPEP trials did not observe a statistically significant increase in doxycycline resistance in N. gonorrhoeae strains although the sample sizes were small and with relatively short follow-up (median of 9 months).^16,18 Similarly, oropharyngeal commensal Neisseria species, which may constitute a risk for AMR through horizontal gene transfer, did not significantly increase between baseline (63%) and follow-up (70%) in the doxycycline PEP arm in the DoxyPEP trial.²² However, the impact of doxycycline PEP on AMR in N. gonorrhoeae may be limited in South Africa as studies report that 74% – 89% of N. gonorrhoeae strains are already resistant to tetracyclines.^12,13,14 Also, hypothetical concerns about co-selection of resistance to other classes of antibiotics may be limited given that most N. gonorrhoeae strains already have the putative tet(M) and rpsJ V57M mutations.^13,14,23

There is little evidence of AMR in C. trachomatis and T. pallidum despite widespread use of doxycycline for these infections.^9,20 Rare reports of treatment failure have been published.²⁰ Resistance could theoretically emerge through single-point mutations or gene transfer, but this is generally considered unlikely.^16,18,20,22 M. genitalium has the potential to acquire tetracycline resistance given that this organism has a history of rapidly acquiring resistance mutations. Despite the low rate (< 2%) of resistance to azithromycin,^24,25 previously as stand-alone treatment, sequential therapy of doxycycline followed by azithromycin or moxifloxacin is now recommended to avoid further emergence of AMR.¹¹ The effects of doxycycline PEP on development of AMR in M. genitalium are unknown but warrant close monitoring.

Treatment with tetracyclines has been associated with AMR in ‘bystander’ respiratory and gastrointestinal tract pathogens.²⁰ Data from a limited number of prospective studies suggest that treatment with oral tetracyclines for 2–18 weeks may increase resistance in the bacterial flora, although the effects may be modest and transient.²⁶ The effects of doxycycline PEP on the microbiome may be different than when used daily for treatment; for example, the median number of pills taken in the DOXYVAC trial was 7 per month (interquartile range [IQR]: 4–11) and 4 per month (IQR: 1–10) in the DoxyPEP trial.^17,18 Nevertheless, the DoxyPEP trial reported a decrease from 42% to 29% in nasopharyngeal carriage of S. aureus in a small sample of doxycycline PEP users, with an increase from 4% to 12% in the proportion of doxycycline-resistant strains between baseline and 1 year follow-up.²² The DoxyPEP study reported a persistently high (> 80%) prevalence of nasopharyngeal commensal Neisseria species in the doxycycline PEP arm, also with a nonsignificant increase in proportion of doxycycline-resistant isolates (63% to 70%).²² Studies with larger sample sizes and longer follow-up time are required to better understand impact of doxycycline PEP on STIs, other pathogens and the microbiome. This could be done by close monitoring and AMR surveillance of indiviudals using doxycycline PEP in health programmes.

Doxycycline is a well-known, well-tolerated antibiotic that has been used for decades in the treatment of infectious diseases.⁷ Randomised controlled trials have shown efficacy of doxycycline PEP for STI prevention in MSM, but not in young cisgender women, while efficacy has not been evaluated in female sex workers and men who are not MSM.^16,17,18 Therefore, consideration of doxycycline PEP for STI prevention should be limited to MSM at this stage.

Incidence of STIs in MSM in South Africa are higher than reported in the trials, suggesting that the number needed to treat to prevent one STI may be smaller than 4.7 in MSM on HIV PrEP and 5.3 in PLHIV, as reported in the DoxyPEP trial.¹⁸ The preventive effect would be predominantly to reduce C. trachomatis and T. pallidum infections due to the high rate of tetracycline resistance in N. gonorrhoeae.^12,13,14 Given the high burden, associated morbidity and effects on HIV transmission, the efficacy of doxycycline PEP, and the limited alternative preventive options available, it would be reasonable to offer doxycycline PEP for primary STI prevention as part of a comprehensive sexual health package to high-risk MSM.

Diagnostic STI testing should be performed before prescribing doxycycline PEP to avoid suboptimal treatment and potentially driving AMR of any undiagnosed prevalent infection. Resources permitting, diagnostic testing should include sensitive nucleic acid amplification tests for oropharyngeal, urethral, and anorectal C. trachomatis and N. gonorrhoeae infection as well as syphilis serology.¹¹ Diagnostic tests for STIs are not routinely available in South Africa, but there are settings where access to diagnostic tests can be organised, for example in the private sector, research studies and implementation demonstration projects. Importantly, access to diagnostic STI testing and biomedical preventions for MSM are key objectives of the WHO’s Global Health Sector strategies on HIV, viral hepatitis and STIs,²⁷ as well as the National Strategic Plan for HIV, tuberculosis and STIs 2023–2028.²⁸

There are many unanswered questions and genuine concerns about the long-term effects of doxycycline PEP on AMR in STIs, other pathogens and the microbiome.²⁰ The impact of doxycycline PEP on AMR in N. gonorrhoeae may be relatively small given that most strains are resistant to tetracyclines.^12,13,14 Emergence of doxycycline resistance in other STIs, especially M. genitalium, would be highly concerning given the limited treatment options available. Therefore, we recommend that doxycycline PEP is offered in well-controlled settings and with access to diagnostic testing (including AMR) by clinicians with expertise in STI management.

The potential effects of doxycyline PEP on AMR in other pathogens and the microbiome should be considered in the South African context. Doxycycline has only a few indications, alternative treatment options are generally available and AMR development is unlikely (e.g., tick bite fever and malaria prophylaxis).¹³ Furthermore, tetracyclines are used widespread as the primary antibiotic in agriculture;²⁹ this may have more impact on driving population-level AMR than the effects of doxycycline PEP use by a small subgroup. On the other hand, intensive use of doxycycline in a small core group of users may select or induce AMR in pathogens that could subsequently spread to the general population.¹⁸ Although no worrisome trends in AMR were observed in the doxycycline PEP trials, further research is warranted to better understand the effects of doxycycline PEP on tetracycline resistance at both the individual and population level.

Globally, the balance between effectiveness of doxycycline PEP and AMR concerns is weighed differently. The San Francisco Public Health Unit, the California Department of Public Health and the Public Health Seattle & King Country in the US recommend doxycycline PEP use by MSM and TGW with a recent history of STI.^30,31,32 In the absence of long-term data regarding the impact of doxycycline PEP on AMR and the microbiome, the Centers for Disease Control and Prevention (CDC) provides considerations to healthcare providers to inform decision-making on a case-by-case basis, and additional guidelines are anticipated later in 2023.³³ The Australasian Society for HIV, Viral Hepatitis and Sexual Health Medicine is planning to convene a forum in 2023 to provide evidence-based guidance and recommendations.⁹ The British Association for Sexual Health and HIV and the United Kingdom (UK) Health Security Agency do not endorse doxycycline for STI prophylaxis for lack of data about AMR, while other countries are waiting for further data before making formal recommendations.³⁴ The WHO has not yet issued any statement.

Doxycycline PrEP (200 mg daily) is being evaluated as an alternative to PEP. One Canadian RCT comparing doxycycline PrEP and doxycycline PEP has started and an Australian study is giving MSM the option of doxycycline PEP or PrEP. These studies follow a pilot RCT that showed a 73% reduction in bacterial STIs in a sample of 30 MSM.³⁵ However, at this stage, there is insufficient evidence to recommend daily doxycycline PrEP.

Doxycyline PEP is a novel and effective preventive intervention that could substantially reduce the high STI burden and associated morbidity in MSM in South Africa. These STI prevention benefits should be weighed against the largely unknown risks and potential future harms of AMR. In the absence of current local doxycycline PEP data, it is essential that South Africa builds expertise in doxycycline PEP provision and improves STI and AMR surveillance. Based on current evidence, we think it is reasonable to offer doxycycline PEP to high-risk MSM on a case-by-case basis following comprehensive sexual health counselling by experienced clinicians in settings that have access to diagnostic STI testing and ongoing AMR surveillance.