What is the best alternate treatment for latent syphilis other than Penicillin g benzathine

Comment by InpharmD Researcher

Available evidence identifies doxycycline as the best-supported alternative to benzathine penicillin G for latent syphilis in nonpregnant adults and adolescents, although noninferiority to penicillin has not been established. CDC guidelines list doxycycline 100 mg orally twice daily or tetracycline 500 mg orally four times daily, each for 28 days, as the only acceptable alternatives for late latent syphilis or syphilis of unknown duration in nonpregnant patients with penicillin allergy, without preferring either regimen. For early latent syphilis, doxycycline 100 mg orally twice daily for 14 days is an alternative. Comparative retrospective studies and meta-analyses found no statistically significant differences in serologic response between doxycycline and benzathine penicillin G; however, low evidence certainty and limited stage-specific reporting restrict definitive conclusions. Of note, alternative regimens require close clinical and serologic follow-up. Additionally, patients with penicillin allergy who are pregnant or whose adherence or follow-up cannot be ensured should undergo desensitization and receive penicillin G.
Background

The CDC’s 2021 Sexually Transmitted Infections Treatment Guidelines identify doxycycline 100 mg orally twice daily or tetracycline 500 mg orally four times daily, each for 28 days, as the only acceptable alternatives to penicillin for late latent syphilis or syphilis of unknown duration in nonpregnant patients with penicillin allergy. The guidelines do not designate either regimen as preferred and note that the effectiveness of alternatives to penicillin for latent syphilis has not been well documented, particularly among persons with HIV infection. These regimens require close clinical and serologic follow-up, with quantitative nontreponemal testing recommended at 6, 12, and 24 months. For clearly defined early latent syphilis, nonpregnant patients with penicillin allergy should respond to alternative antibiotics recommended for primary and secondary syphilis. Ceftriaxone might be effective for latent syphilis; however, its optimal dose and duration remain undefined, and treatment decisions should involve specialist consultation. Patients with penicillin allergy whose adherence or follow-up cannot be ensured, as well as pregnant patients with penicillin allergy, should undergo desensitization and receive penicillin G. [1]

A 2024 meta-analysis evaluated nonpenicillin treatment strategies for nonneurological syphilis across 27 comparative studies involving 6,710 patients, including patients with early and late latent syphilis. Searches included six databases from inception through August 28, 2023; the primary outcome was serological cure, most commonly defined as a ≥4-fold decline in nontreponemal antibody titers, although assessment time points varied. Across nine studies comparing doxycycline with benzathine penicillin G, serological cure rates did not differ significantly (pooled odds ratio [OR] 0.82; 95% confidence interval [CI] 0.61 to 1.10; I²= 1%). Doxycycline regimens generally consisted of oral doses of ≥100 mg twice daily for 14 to 28 days, depending on disease stage. Similarly, five studies comparing ceftriaxone with benzathine penicillin G found no significant difference in serological cure (pooled OR 1.66; 95% CI 0.97 to 2.84; I²= 0%). Although azithromycin also demonstrated no significant difference from penicillin, the authors acknowledged that CDC guidance discourages its use because of antimicrobial resistance concerns. Limitations included predominantly nonrandomized studies (17/27), variable treatment regimens and definitions of serological cure, and incomplete adverse event and reinfection data. These pooled comparisons encompassed multiple syphilis stages and did not establish a preferred alternative specifically for latent syphilis. [2]

A 2026 network meta-analysis evaluated alternative antibiotic regimens for early and late syphilis, including 14 randomized controlled trials (RCTs; N= 2,280) and 20 observational studies (N= 6,895). Thirteen RCTs involving 1,623 participants and 10 treatments contributed to the network meta-analysis, which assessed cure at 12 months or the nearest reported time point; all studies incorporated serologic response into their cure definitions. Doxycycline monotherapy did not demonstrate a statistically significant difference in cure compared with benzathine penicillin G (OR 6.88; 95% credible interval [CrI] 0.14 to 625.93), although the point estimate favored doxycycline. Observational meta-analyses likewise found no statistically significant differences between benzathine penicillin and doxycycline or other tetracyclines. The review described doxycycline 100 mg orally twice daily for 14 days as a recommended alternative for early latent syphilis and a 28-day course for late latent syphilis. However, insufficient reporting of syphilis stage prevented the planned stage-specific analyses, and evidence certainty was generally low for RCTs and very low for observational studies because of risk of bias and imprecision; consequently, noninferiority of alternative regimens could not be established. The authors concluded that doxycycline remained the best-supported alternative to benzathine penicillin for early and late latent syphilis in nonpregnant adults and adolescents, while acknowledging potential adherence challenges and the limitations of the available evidence. [3]

The 2016 World Health Organization (WHO) guidelines, which remain applicable to syphilis treatment in nonpregnant adults and adolescents, distinguish early latent syphilis (infection of no more than 2 years’ duration) from late latent syphilis or infection of unknown duration. For early latent syphilis, WHO recommends benzathine penicillin G 2.4 million units intramuscularly once. When benzathine or procaine penicillin cannot be used or is unavailable, WHO suggests doxycycline 100 mg orally twice daily for 14 days or ceftriaxone 1 g intramuscularly daily for 10 to 14 days; doxycycline is preferred over ceftriaxone because of its lower cost and oral administration. Azithromycin 2 g orally once is suggested only in special circumstances when local susceptibility is likely. For late latent syphilis or infection of unknown duration, WHO recommends benzathine penicillin G 2.4 million units intramuscularly weekly for 3 consecutive weeks. When penicillin cannot be used, including allergy for which desensitization is not possible, or is unavailable, WHO suggests doxycycline 100 mg orally twice daily for 30 days. These alternative-treatment recommendations are conditional and supported by very low-quality evidence. [4]

Background References: [1] Centers for Disease Control and Prevention. Latent syphilis. Sexually Transmitted Infections Treatment Guidelines, 2021. Reviewed July 22, 2021. Accessed October 29, 2026.
[2] Callado GY, Gutfreund MC, Pardo I, et al. Syphilis treatment: systematic review and meta-analysis investigating nonpenicillin therapeutic strategies. Open Forum Infect Dis. 2024;11(4):ofae142. doi:10.1093/ofid/ofae142.
[3] Cvetkovic A, Erdman L, Johnstone R, Bartoszko JJ, Turner S, Mertz D. Effective Alternative Treatment Regimens for People With Infectious Syphilis: A Systematic Review and Network Meta-Analysis. Sex Transm Dis. 2026;53(9):626-636. doi:10.1097/OLQ.0000000000002341
[4] World Health Organization. WHO Guidelines for the Treatment of Treponema pallidum (Syphilis). World Health Organization; 2016. Accessed October 29, 2026. https://www.who.int/publications/i/item/9789241549714
Literature Review

A search of the published medical literature revealed 3 studies investigating the researchable question:

What is the best alternate treatment for latent syphilis other than Penicillin g benzathine

Level of evidence

B - One high-quality study or multiple studies with limitations  Read more→



Please see Tables 1-3 for your response.


Is 3 Really What We Need? Relative Effectiveness of Benzathine Penicillin G and Doxycycline Treatment Regimens for Late or Unknown Duration Syphilis in 6 US Jurisdictions, 2016–2021

Design

Retrospective cohort study

N= 18,027

Objective

To compare the treatment effectiveness of 1 versus 3 doses of benzathine penicillin G (BPG) and doxycycline for late or unknown duration syphilis

Study Groups

3 doses of BPG (n= 11,342)

1 dose of BPG (n= 1,928)

Doxycycline (n= 3,017)

Inclusion Criteria

Cases diagnosed with late or unknown duration syphilis between 2016 and 2021 from 6 US jurisdictions with recorded treatment and follow-up titers

Exclusion Criteria

Cases with missing baseline or follow-up titers, treatment, or those who received an alternative treatment regimen

Methods

Routinely collected surveillance records provided demographic characteristics, treatment medications/doses/dates, and quantitative nontreponemal titers. BPG regimens comprised 2.4 million units IM per dose and were classified by dose number and interval; permissible intervals were 5–14 days. Doxycycline regimens were 100 mg twice daily, categorized as <28 or ≥28 days, with the latter including multiple courses totaling >28 days. Baseline was the most recent titer within 90 days before treatment, and response was assessed using rapid plasma reagin, venereal disease research laboratory, or unheated serum reagin tests. Follow-up titers were censored at reinfection or retreatment occurring >90 days after treatment initiation. Log-binomial regression compared response with BPG 3 doses at permissible intervals as the reference, adjusting for jurisdiction, age, sex/sex of partners, pregnancy status, and HIV status; analyses were also stratified by baseline titer.

Duration

2016 to 2021

Outcome Measures

Primary: 4-fold decrease in nontreponemal serum titer by 24 months posttreatment

Baseline Characteristics

 

All patients (n= 18,028)

Mean age, years

36.2 ± 12.8

Female

4,598 (26%)

HIV positive

6,675 (37%)

Pregnant females

1,404 (31%)

White, non-Hispanic

4,446 (25%)

Black, non-Hispanic

6,811 (38%)

Hispanic/Latino

5,066 (28%)
Results

≥4-Fold Titer Decrease

Doxycycline ≥28 Days (n= 3,017) BPG 3 Doses, 5–14-Day Intervals (n= 4,710) BPG 3 Doses, Unknown Intervals (n= 6,632)

By 6 months

1,299 (43%) 2,095 (44%) 2,889 (44%)

By 9 months

1,588 (53%) 2,525 (54%) 3,533 (53%)

By 12 months

1,808 (60%) 2,828 (60%) 4,058 (61%)

By 24 months, overall

2,240 (74%) 3,414 (72%) 5,063 (76%)

The authors reported no statistically significant difference in overall treatment effectiveness between doxycycline and BPG regimens and no evidence of effect modification by HIV or pregnancy status.

In a restricted analysis of nonpregnant females of reproductive age, response was 72% with doxycycline versus 77% and 80% with 1 and 3 BPG doses, respectively; detailed estimates and p-values were not provided.

Adverse Events

No specific adverse events reported in the study.

Study Author Conclusions

We did not observe a difference in treatment effectiveness between 1 and 3 doses of BPG, or between 3 doses of BPG and doxycycline, for late or unknown duration syphilis. Additional observational and clinical studies with more complete treatment and serologic response information are needed to confirm these findings and inform practice recommendations.

Critique

The large cohort across 6 jurisdictions and adjusted, titer-stratified analyses provide useful comparative evidence for doxycycline; however, exclusion of approximately two-thirds of reported cases, incomplete treatment/adherence documentation, and possible misclassification of early infection limit causal interpretation. The findings support doxycycline as an alternative associated with similar serologic response, but do not establish equivalence, comparative safety, prevention of late complications, or superiority over other alternatives.

Table 1 References:
[5] Pugsley RA, Davis NL, Cope AB, et al. Is 3 Really What We Need? Relative Effectiveness of Benzathine Penicillin G and Doxycycline Treatment Regimens for Late or Unknown Duration Syphilis in 6 US Jurisdictions, 2016-2021. Clin Infect Dis. 2026;82(4):e843-e851. doi:10.1093/cid/ciag099

Comparison of Three Treatment Regimens for Latent Syphilis of Late or Unknown Duration Using Public Health Surveillance Data

Design

Retrospective cohort study

N= 761

Objective

To compare rates of serological cure among persons with latent syphilis of late or unknown duration (LSUD) following receipt of one of three regimens: one dose of benzathine penicillin G (BPG), three doses of BPG, and twenty-eight days of doxycycline

Study Groups

Single-dose BPG (n= 54)

Three-dose BPG (n= 611)

Doxycycline (n= 96)

Inclusion Criteria

Persons diagnosed with LSUD between 2007 and 2020 with an initial rapid plasma reagin (RPR) ≥1:2 and follow-up RPR testing 1–36 months post-treatment

Exclusion Criteria

Persons without follow-up RPR in Washington state within 3 years of treatment; RPR 1:1 or nonreactive; received 2 BPG injections or fewer than 28 days of doxycycline

Methods

Investigators used surveillance records containing syphilis staging, treatment information, and reported RPR results to compare one intramuscular BPG dose, three weekly intramuscular BPG doses, and doxycycline 100 mg orally twice daily for 28 days. Analyses were stratified by initial RPR titer as high (≥1:32) or low (<1:32), and serological cure was defined as a ≥4-fold decline from the initial RPR titer. Cox proportional hazards models adjusted for initial RPR titer; follow-up ended at the first qualifying decline or the last reported RPR within 36 months for those without cure. Sensitivity analyses examined persons with initial follow-up within 180 days and used inverse probability weighting incorporating initial RPR titer, HIV status, race, ethnicity, and age; weighting was limited to the high-titer population because few low-titer patients received single-dose BPG.

Duration

2007 to 2020

Outcome Measures

Serological cure, defined as a ≥4-fold (two-titer) RPR decline, overall and stratified by initial RPR titer; time to serological cure

Baseline Characteristics  

All patients (n= 761)

Median age, years (IQR)

35 (28 to 47)

Cisgender women

16.8%

Cisgender MSM

62%

Race

White

Black

Hispanic/Latinx

 

49%

15%

18%

Living with HIV

38%

Prior syphilis

19%

Unknown HIV status

22%

Initial RPR ≥1:32

63.5%

Abbreviations: IQR, interquartile range; MSM, men who have sex with men

Results  

Single-dose BPG

Three-dose BPG Doxycycline p-value
Serological cure, all initial titers

40 (74%)

444 (73%) 72 (75%) -
High-titer serological cure

36 (88%)

330 (88%) 58 (88%) 1.00
Low-titer serological cure

4 (31%)

114 (49%) 14 (47%) 0.458

After adjustment for initial RPR titer, time to serological cure did not differ significantly between one-dose and three-dose BPG in high-titer (hazard ratio [HR] 0.89; 95% confidence interval [CI] 0.63 to 1.26) or low-titer patients (HR 0.69; 95% CI 0.25 to 1.86), or between one-dose BPG and doxycycline in either subgroup (HR 1.03; 95% CI 0.68 to 1.58; and HR 0.94; 95% CI 0.28 to 3.13, respectively). Sensitivity analyses yielded similar findings. A direct doxycycline-versus-three-dose BPG HR was not reported.

Adverse Events

Not reported

Study Author Conclusions

In conclusion, we found that 1 dose of BPG, 3 doses of BPG, and 28 days of doxycycline were associated with identical levels of serological response among patients with LSUD and an RPR ≥1:32.

Critique

The cohort directly compared 28-day doxycycline with BPG regimens and incorporated RPR stratification and adjusted sensitivity analyses, but nonrandom treatment selection, exclusion of patients without follow-up, and use of serological response rather than microbiological cure limit causal interpretation. Similar observed response rates support doxycycline as an alternative evaluated in this study, but do not establish equivalence or identify the best alternative, particularly for confirmed late latent infection, which could not be meaningfully separated from infection of unknown duration.

Table 2 References:
[6] Hunt TJ, Berzkalns A, Cannon CA, Menza TW, Dombrowski JC, Golden MR. Comparison of 3 Treatment Regimens for Latent Syphilis of Late or Unknown Duration Using Public Health Surveillance Data. Sex Transm Dis. 2025;52(8):457-461. doi:10.1097/OLQ.0000000000002148

Late Latent Syphilis in a Patient with End-Stage Renal Disease and Presumptive Penicillin Allergy

Design

Case report

History of Present Illness

A 58-year-old white female patient was admitted to the hospital with altered mental status, shortness of breath, chest pain, increased sputum production, and cough.

Past Medical History

Her medical history included ESRD requiring hemodialysis three days a week, chronic obstructive pulmonary disease, type II diabetes, hyperlipidemia, neuropathy, muscle weakness, seizure, and acute bacterial endocarditis secondary to mitral valve replacement two months before.

She had a documented penicillin allergy of facial swelling.

Case Presentation

The patient tested positive for the presence of syphilis immunoglobulin G (IgG) antibody, had a nonreactive nontreponemal test (the RPR) and a reactive treponemal test (the FTA-ABS), confirming the diagnosis of syphilis. The patient was diagnosed with late latent syphilis of unknown duration, and a standard course of penicillin G benzathine was prescribed. 

The patient received oral penicillin desensitization per hospital policy, subsequently completing the full course of penicillin G therapy with no reported reactions. The desensitization protocol involved 14 oral doses of penicillin V potassium starting at 0.0625 mg (100 units of penicillin V) and ending at 400 mg (640,000 units of penicillin V) every 15 minutes. After completion of this protocol, the first dose of intramuscular penicillin G benzathine 2.4 million units intramuscularly. 

Study Author Conclusions

After the completion of an oral desensitization protocol, the three-dose standard regimen of intramuscular penicillin G for late latent syphilis was safely administered to a hemodialysis patient without dosage adjustment or repeated desensitization.

Table 3 References:
[7] Grice KT, Day SA. Late latent syphilis in a patient with end-stage renal disease and presumptive penicillin allergy. Am J Health Syst Pharm. 2014;71(7):558-61.