Introduction
Syphilis
Syphilis is a systemic bacterial disease caused by Treponema pallidum subspecies pallidum. The risk of transmission after sexual intercourse is estimated to be approximately one third. Primary syphilis typically manifests as a chancre (genital ulcer) around 3 weeks after exposure. If left untreated, the lesion may heal, but 9 to 16 weeks after exposure, symptoms of secondary syphilis may appear, usually as mucocutaneous lesions and non-specific symptoms such as fever and malaise. These symptoms may resolve even without treatment, leading to an asymptomatic but still infectious stage of the disease – latent syphilis. Within 1 year after infection, the disease is classified as early syphilis, and beyond 1 year as late syphilis. Late syphilis can be asymptomatic (latent late syphilis) or manifest as tertiary syphilis with varying symptoms depending on the affected organ (e.g., cardiovascular syphilis). Central nervous system (CNS) involvement, known as neurosyphilis, can occur at any stage of the infection [1, 2].
Humankind is the sole hosts of syphilis. So far, no effective method of culturing T. pallidum in vitro has been developed [3, 4]. Therefore, the diagnosis is established based on serologic tests and the clinical symptoms. The diagnosis requires a positive result of at least one non-treponemal (non-specific, e.g. rapid plasma reagin (RPR), venereal disease research laboratory (VDRL)) and one treponemal (specific, e.g. Treponema pallidum hemagglutination assay (TPHA)) test [1].
The treatment of the first choice for early syphilis is a single intramuscular dose of benzathine penicillin of 2.4 million IU. Whereas, in late syphilis, benzathine penicillin is administered at a dose of 2.4 million IU i.m. once a week for the 3 consecutive weeks. A proper response to treatment is based on the resolution of symptoms of the disease and at least a fourfold decrease in the titre of non-treponemal assays after 6 months from the end of antibiotic therapy compared to pre-treatment results [2].
Serofast state in syphilis
Serofast state is defined by the Centers for Disease Control and Prevention (CDC) as an insufficient (i.e. less than at least 4-fold) decrease in the non-treponemal titre after 6–12 months for early syphilis or after 12–24 months for late syphilis since the recommended treatment was administered, with concomitant resolution of symptoms of infection [5]. Serofast state affects 15–20% (from 5% to up to 41%) of patients after recommended treatment [1, 6, 7]. So far, the mechanism of action and the clinical significance of serofast state in syphilis patients have not been sufficiently elucidated. The constant production of antibodies expressed in non-decreasing titres of non-treponemal assays may be due to the survival of live treponemal in privileged areas of the body e.g. the CNS, but may also be associated with over-stimulation of the patient’s immune system, despite successful eradication of the bacteria. Therefore, it remains unclear whether a patient with a serofast state can be considered cured and how best to manage such patients in daily medical practice [6].
Aim
The aim of this article is to critically evaluate the current scientific understanding of the serofast state in patients with syphilis and to highlight the issue of the unclear pathomechanism of this phenomenon and, most importantly, the treatment of this group of patients.
Review material and methods
Epidemiological data indicate that the incidence of syphilis is on the rise, and the number of cases showing a serofast state after treatment is also increasing [2, 3, 6, 8]. This suggests that we are dealing with a growing group of patients for whom recommendations are not clearly defined.
The PubMed database was searched for information on the serofast state in syphilis, and the search results were critically analysed. The review was limited to publications from the period of 2010–2023. The databases were searched in May, June, and July 2023.
After reading and critically analysing the retrieved publications, 33 papers were selected for inclusion in this review. The remaining publications were excluded by the authors as they did not contribute relevant data or were too distant from the topic of this article. The review also considered the conclusions of other review articles on similar topics and the CDC guidelines, which, in the authors’ opinion, broaden the perspective and enhance the value of the publication.
Based on the available literature, the main research questions concerning the phenomenon of the serofast state in syphilis were formulated:
Results and discussion
Based on the publications included in the review, the following sub-topics of serofast state addressed in the literature were highlighted:
Predictive factors of serofast state
A significant proportion of the retrieved publications are original studies related to predictive factors of the serofast state or curability.
The initial observation that emerges from an analysis of the literature is that lower titres in non-treponemal assays at the time of diagnosis are associated with smaller declines in these titres after treatment. Moreover, a strong correlation was found between early latent syphilis and risk of serofast state after syphilis treatment when compared to patients with primary and secondary syphilis [7, 9–17].
The baseline titres are correlating with the disease activity and the number of treponema host organism. In addition, the pro-inflammatory response to Treponema pallidum infection is weakening with the duration of the disease. It has been shown that in the later stages of infection, the regulatory response starts to dominate. This partially explains the difficulties in treating late syphilis [7, 11, 13, 17, 18]. These observations also seem to explain the higher incidence of serofast syphilis in patients treated for both early and late latent syphilis and in patients whose initial titres of non-treponemal reactions were low [7, 10–13, 19–21].
Following the administration of antibiotics, individuals with syphilis may develop a Jarisch-Herxheimer reaction. It results from the release of toxins from the antibiotic-killed Treponema pallidum and occurs 1–1.5 h after antibiotic administration and manifests clinically with symptoms such as fever, chills, myalgias, headache and exacerbation of cutaneous lesions. Studies have shown that the occurrence of a Jarisch-Herxheimer response is a predictor of a proper serological response to syphilis treatment. This confirms observations regarding the important role of the host immune response in the pathogenesis of the serofast state syphilis [11].
Data indicate also that the risk of developing serofast state after syphilis treatment increases with age. This also appears to be related to the observed weakening of the pro-inflammatory component of the immune system with age [7, 10–12, 17].
Subsequent studies have demonstrated that the serofast state syphilis is more prevalent in women and in patients who report a high number of sexual partners in the 6 months prior to contracting the disease. The precise pathomechanism underlying these observations remains unknown [5, 10–12].
The correlation of HIV infection with proper response to therapy could not be clearly determined. Two studies showed that patients with HIV infection were less likely to have a proper response to treatment, however as many as seven studies showed that HIV status did not affect treatment response. In some studies, a reduced CD4 cell count was associated with an increased risk of treatment failure, but other studies did not support this correlation [17].
The intense pro-inflammatory response of the body to Treponema pallidum infection expressed by an increase in pro-inflammatory cytokines (TNF-a, IFN-g, IL-6) and C-reactive protein (CRP) is associated with a greater likelihood of a proper response to treatment [16].
As it turns out the presence of non-specific antinuclear antibodies (ANA) in the blood of patients with syphilis as well as a weak humoral response against a specific TpN47 treponemal antigen may prove to be valuable predictors of serofast state, but further studies on a larger population are necessary to confirm these conclusions [6, 20].
The data also lead to the conclusion that serofast state does not change significantly between follow-up examinations at 6 and 12 months [22].
There are reports indicating that untreated HIV infection may increase the likelihood of serofast state syphilis. Data on the effect of viral load and CD4+ cell count on the occurrence of serofast state in syphilis are divergent [6, 23]. It is noteworthy that because of the prolonged serological response time to syphilis treatment in patients with HIV co-infection, serofast state should only be assessed at 12 months rather than 6 months after treatment [9, 22, 23].
Evaluation of the cytokine profile and microRNA expression profile in syphilis, may prove to be a breakthrough in the diagnosis and prediction of serofast state, but studies on a larger group are needed [19, 24–28].
Antibiotic therapy and serofast state
Several studies showed that the use of a three-dose rather than a single-dose benzathine penicillin treatment regimen for the first-line therapy does not reduce the rate of serofast state in early syphilis [5, 9, 14, 22]. Only a few studies indicated that the use of azithromycin or erythromycin in treatment of syphilis may be associated with the serofast state [7, 10]. However, these conclusions are based on small-population studies.
Re-treatment with penicillin in serofast state patients is not an effective treatment for serofast state conditions in a single-dose regimen [9, 15, 20, 29], but in a three-dose regimen, it may provide therapeutic benefit in almost half of patients [30]. However, the CDC stresses that there are still too few studies on this issue to draw firm conclusions.
The finding of the lack of response of most subjects with serofast state syphilis to re-treatment can be explained by the hypothesis that serofast state in this group is not due to persistent treponemal infection, but to a persistent, excessive immune response to the T. pallidum antigen, resulting in the continuous production of antibodies despite the absence of a live pathogen [6, 20].
Neurosyphilis and serofast state
In a study involving 402 HIV-negative patients diagnosed with serofast state syphilis, it was found that 34.6% had asymptomatic neurosyphilis (ANS). This represents a significantly higher proportion of ANS compared to patients with latent syphilis or syphilis alone, where ANS occurs in only 13.5% of cases. Additionally, about 35% of patients who were initially diagnosed with ANS went on to develop symptomatic neurosyphilis. These findings highlight a notably higher risk of neurosyphilis among those in a serofast state, suggesting that serofast state syphilis may be more closely linked to the progression of neurological involvement than previously thought [31].
Another study found that of 324 HIV-negative patients with serofast state syphilis, 27.5% had ANS. The study identified the following risk factors for ANS: less than fourfold decrease in RPR reactions after treatment, current titre of serum RPR test of more than 1: 32 [32].
According to studies on smaller groups of individuals, the association between serofast state and asymptomatic neurosyphilis varies depending on HIV status. 30.7% of HIV-negative patients with serofast state were diagnosed with ANS compared to 41.7% who were serofast but HIV positive. These findings suggest a significant correlation between serofast state and the presence of asymptomatic neurosyphilis, particularly in patients with HIV co-infection [6, 16, 23].
A Chinese meta-analysis examining the prevalence of asymptomatic neurosyphilis (ANS) in HIV-negative patients with serofast state syphilis reported an overall ANS rate of 13%. However, the analysis faced challenges due to the wide variation in results, which largely depended on the definition of ANS used. When using the stricter definition of “verified ANS”, characterized by reactive VDRL reactions in the cerebrospinal fluid (CSF), the prevalence was much lower, at only 3%. In contrast, when ANS was defined more broadly as “probable ANS”, based on elevated white blood cell (WBC) counts and protein levels in the CSF, the prevalence increased significantly to 21%. This discrepancy highlights the impact of diagnostic criteria on the reported prevalence of ANS in serofast state syphilis patients [33].
A study on patients with serofast state syphilis and HIV co-infection revealed that those who presented with asymptomatic neurosyphilis (ANS) had lower CD4+ cell counts (indicating higher levels of immunosuppression) and higher HIV viral load compared to both serofast state patients without ANS and HIV-positive patients without serofast state. Notably, none of the patients who were receiving antiretroviral therapy (ART) developed asymptomatic ANS. This finding strongly suggests that ART significantly reduces the risk of developing asymptomatic neurosyphilis in HIV-positive individuals [23].
Based on the available studies [16, 23, 31], it is advisable to consider cerebrospinal fluid (CSF) examination in patients with serofast state syphilis. Early diagnosis and treatment of asymptomatic neurosyphilis could help reduce the risk of progression to the symptomatic stage of CNS involvement.
According to current CDC recommendations for CSF testing:
– CSF testing should be performed in all individuals with neurological symptoms and positive syphilis serological reactions [34].
– Routine CSF testing is not recommended in individuals with syphilis [35].
– CSF testing after syphilis treatment is not necessary in immunocompetent individuals (defined as HIV-negative persons and HIV-positive persons effectively treated with antiretroviral therapy) if serum serological tests show a proper decline and clinical symptoms resolve [5, 36].
However, in cases where there is a lack of proper serological response to syphilis treatment or in immunocompromised patients, additional CSF testing may be warranted to ensure adequate management and to detect potential complications.
Pregnancy and serofast state
Treatment of syphilis in pregnancy with benzathine penicillin is a recommended and safe treatment for mother and child [37–39].
Patients who have undergone treatment for syphilis prior to pregnancy and are found to be serofast state with stable, low titres of non-treponemal assays may not require treatment, but an antibiotic should be included if titres rise [39].
Studies on a small group of pregnant syphilis patients indicate a higher incidence of serofast state at 6 months after treatment in this population (42.9%) than in the general population. However, in all patients with serofast state, reaction titres already achieved an expected at least fourfold decrease at 9 months after antibiotic treatment, leading to the conclusion that in pregnant patients, non-treponemal titres may decrease more slowly after penicillin treatment than in nonpregnant patients [37].
Patients with syphilis during pregnancy had statistically significantly higher concentrations of IL-10 and TGF-b, indicating a predominant regulatory immune response in these women. Pregnant patients with an abnormal serological response to treatment compared to those with a good response also showed higher concentrations of these cytokines. The older the pregnancy when penicillin treatment was initiated, the longer the time to achieve a proper serological response [37].
Patients with serofast state in pregnancy should have their non-treponemal titres checked monthly to assess the dynamics of reaction changes and to exclude syphilis reinfection [38].
Position of Centers for Disease Control and Prevention (CDC)
The CDC regularly publishes guidelines for the treatment of syphilis, and several key issues related to serofast state syphilis are highlighted below:
Treatment regimen consistency: The CDC advises that using more than one dose of benzathine penicillin for treating early and late latent syphilis in patients with HIV co-infection (as well as those without HIV co-infection) does not provide additional clinical or serological benefits. Therefore, the treatment regimen should remain the same regardless of HIV status [5, 36].
Alternative antibiotic options: When penicillin cannot be used, doxycycline is considered an acceptable alternative. However, azithromycin should not be used due to widespread macrolide resistance in the USA and many other regions. Ceftriaxone is also regarded as a reasonable alternative to penicillin for treating early syphilis [5].
Insufficient data on additional antibiotic doses: There are currently insufficient data on the effectiveness of additional doses of antibiotics in improving the serological response in patients with serofast state syphilis, particularly concerning long-term outcomes. The CDC also acknowledges the discrepancies in the definitions of ‘serofast state’ across different studies, which complicates the interpretation of results [5]. This issue has also been noted by the authors of the present review.
Predictive factors: The predictive factors for serofast state syphilis identified by the CDC align with those identified in this review, further validating the conclusions drawn.
These CDC guidelines provide a clear framework for managing syphilis, particularly in the context of serofast state, while also highlighting areas where further research and data are needed.
Conclusions
The precise nature of serofast remains unclear, and further research is required to elucidate this. A deeper understanding of the mechanism will facilitate the development of standards for diagnostic, therapeutic and epidemiological management.
The authors of this publication agree on this point with Lorenz et al. [40]. It is necessary to introduce uniform definitions, criteria and nomenclature in the field of serofast state syphilis so that in future it will be easier to compare the results of the work of different research teams.
