Comparison of sitafloxacin monotherapy versus doxycycline plus rifampicin for acute uncomplicated human brucellosis: a multicenter, randomized, open-label, non-inferiority trial

Summary

Background

Sitafloxacin has demonstrated superior in vitro activity against Brucella species. However, its potential for treating human brucellosis has not been investigated yet. Therefore, this study evaluated and compared the effects of definitive treatment with sitafloxacin monotherapy to those of doxycycline–rifampicin in patients with acute uncomplicated brucellosis.

Methods

This multicenter, open-label, randomized, non-inferiority clinical trial was conducted from August 2023 to July 2025 at three hospitals in China and included 238 adult patients with brucellosis symptoms for less than 3 months who met the study criteria. They were randomized in a 1:1 ratio to receive sitafloxacin monotherapy (100 mg once daily) or combined doxycycline (100 mg every 12 h) and rifampicin (10 mg/kg daily) for 6 weeks. Adherence, efficacy, and safety were evaluated over a 12-month follow-up period. The primary outcome was the clinical response rate at the end of 6 weeks, and the secondary outcomes were treatment failure rate, relapse during follow-up, and safety profiles. The non-inferiority margin was 12%.

Findings

In the intention-to-treat population, the clinical response rate was 89.9% in the sitafloxacin group and 83.2% in the doxycycline–rifampicin group (risk difference [RD] = 6.7%; 95% CI: −2.0 to 15.7). In the per-protocol population, the response rates were 97.3% and 96.1%, respectively (RD, 1.2%; 95% CI: −4.4 to 7.2). The symptom relapse rates were 2.7% and 1.9% in the sitafloxacin and doxycycline–rifampicin groups, respectively (RD, 0.8%; 95% CI: –4.4 to 6.0). The adverse event-related treatment discontinuation rate was numerically lower in the sitafloxacin group (2.5% vs. 6.7%, RD, –4.20%; 95% CI: –10.5 to 1.3), although the difference did not reach statistical significance. Similarly, the overall incidence of treatment-related adverse events was also numerically lower in the sitafloxacin group (10.9% vs. 19.3%; p = 0.07). These results met the prespecified criterion for non-inferiority.

Interpretation

Sitafloxacin monotherapy was non-inferior to doxycycline–rifampicin for treating acute uncomplicated brucellosis, and showed a favorable safety and tolerability profile. Thus, sitafloxacin monotherapy may have potential as an alternative regimen in this clinical setting.

Funding

This work was supported by the National Key Research and Development Program of China (2023YFC2308800)and the Natural Science Foundation of Shandong Province (ZR2023MH169).

Trial registration

ChiCTR2300072703.

Research in context.

Evidence before this study

PubMed and Embase were searched from their inception to October 20, 2025, with no language restrictions. Paper with the search terms “uncomplicated human brucellosis” or “monotherapy” in the title were. Very few comparative studies were found that assess monotherapy, and they included only small numbers of patients. One study compared monotherapy with co-trimoxazole for 45 d vs. combined tetracycline for 21 d and streptomycin for the first 14 d. Both antibiotic regimens initially controlled the signs and symptoms of disease in all patients. The relapse rate was 46.6% in the co-trimoxazole group vs. 14.8% in the tetracycline-streptomycin group (p < 0.05). In another study, 64 patients received cotrimoxazole for 6 months, with an 81.25% cure rate. In over 18% of the patients, a change of treatment was required due to side effects (3.1%), treatment failure, non-completion (12%), or relapse (3.1%). The only prospective, comparative, and randomized study with quinolone monotherapy was published in 1990 by Lang et al. It was a small study on 6 patients with brucellosis treated with ciprofloxacin for 42 d. A comparison was established with 4 other patients treated with doxycycline and rifampicin. Although all patients responded rapidly, five of the six patients receiving ciprofloxacin relapsed following cessation of therapy. Although classic combination regimens can be used to treat brucellosis, convenience of monotherapy may lead to better compliance, which may be critical for a disease that requires weeks of treatment. Sitafloxacin has high oral bioavailability, tissue concentrations, intracellular penetration capacity (fluoroquinolones appear to achieve intracellular concentrations in phagocytic cells that are significantly higher than extracellular concentrations), and in vitro activity against Brucella spp. However, no study has investigated in vitro susceptibility to sitafloxacin in Brucella isolates from Chinese patients. Moreover, the clinical potential of sitafloxacin for brucellosis treatment has not been examined.

Added value of this study

For all 30 Brucella isolates from Chinese patients, sitafloxacin exhibited potent activity. The efficacy of sitafloxacin monotherapy was non-inferior to that of doxycycline plus rifampicin regime as a first-line anti-brucellosis treatment, as determined by intention-to-treat or per-protocol analyses. Both regimens achieved optimal clinical response rates and relapse rates. In the present study, the risk difference for treatment-related adverse events numerically favored sitafloxacin.

Implications of all the available evidence

Sitafloxacin monotherapy was associated with numerically lower rates of specific adverse events. Accordingly, sitafloxacin monotherapy may be a potential alternative regimen for the treatment of acute uncomplicated brucellosis.

Introduction

Brucellosis is a widespread zoonotic disease with global distribution, with more than 170 countries and regions reporting outbreaks in both humans and animals. It is estimated that approximately 2.1 million new human cases occur annually.^1,2 In recent years, China has experienced a high incidence of brucellosis, with the number of reported cases exceeding 60,000 annually for four consecutive years and showing a tendency toward geographic expansion.³ Brucellosis is a systemic infection caused by the genus Brucella which affecting various species of animals and humans.³ Humans are accidental hosts infected by direct or indirect contact with infected animals or contaminated animal products.⁴ Human brucellosis manifests a wide range of clinical manifestations as fever, arthralgia, sweats, fatigue with acute, and chronic disease stages.^3,4 The disease is characterized by relapses, chronicity and sequelae; moreover, its potential for mortality underscores the seriousness of brucellosis as a threat and emphasize the need for appropriate and impactful measures for its treatment and control.5, 6, 7, 8, 9, 10, 11, 12

Risks associated with delayed or improper therapy often leads to a chronic form of brucellosis that has significant health consequences for patients and places a substantial economic burden on families and society.8, 9, 10, 11, 12 However, there is no effective human vaccine, and it is challenging to eradicate brucellosis in animals globally.8, 9, 10 As a result, antimicrobial treatment remains the primary strategy for addressing human brucellosis.^12,13 Because Brucellae are facultative intracellular parasites, it is important to treat patients with drugs that penetrate macrophages.¹³ In 1986, the World Health Organization recommended doxycycline–streptomycin combination therapy and doxycycline–rifampicin combination therapy as the treatments of choice in human uncomplicated brucellosis.¹² Each of these regimens has disadvantages: the relapse rates and the toxicity and side effects.^14,15 On the other hand, the use of rifampicin in areas where brucellosis and tuberculosis are endemic has raised concerns about the development of community resistance to rifampicin.¹⁶ In addition, twice-daily long-term dosing of oral doxycycline plus intramuscular administration of streptomycin often leads to poor patient compliance.¹⁵

Interruption of treatment and poor compliance are important factors that contribute to the ineffectiveness of current brucellosis treatments and its recurrence.¹³ Accordingly, it is important to design treatment regimens with good adherence and few side effects.¹³ A promising agent that has emerged in this regard is the broad-spectrum antimicrobial drug sitafloxacin, which is a 4th generation fluoroquinolone antibiotic with broad activity against a wide range of Gram-negative and Gram-positive bacteria.¹⁷ Fluoroquinolones, as a class, exhibit a broad spectrum of antibacterial activity and are known for their oral bioavailability, high tissue concentrations, ability for intracellular penetration, and in vitro activity against Brucella spp.^17,18 Certain quinolones have shown antibacterial activity against Brucella when administered in combination with other antibacterials. For example, rifampicin with quinolones was found to a high safety profile and good efficacy,¹¹ and the combination of ofloxacin and rifampin administered for 6 weeks was found to be as effective as combined doxycycline–rifampin administered for the same period.¹⁹ Compared to other fluoroquinolones, sitafloxacin displays a lower minimal inhibitory concentration (MIC) for many bacterial species and stronger activity against anaerobes, intracellular bacteria, and persisters.^17,18 Further, in several countries, its use has been approved for the treatment of pneumonia and urinary tract infections, as well as other upper and lower respiratory infections, genitourinary infections, oral infections, and otitis media.^17,20,21 In fact, sitafloxacin exhibited the strongest activity against Brucella in comparison to other tested fluoroquinolones in a previous in vitro study.¹⁷ However, no study has investigated in vitro susceptibility to sitafloxacin in Brucella isolates from Chinese patients. Moreover, the clinical potential of sitafloxacin for brucellosis treatment has not been examined either.

Here, we have expanded on previous studies that have examined the therapeutic effects of sitafloxacin by investigating its effect on brucellosis under in vitro conditions and conducting an investigator-initiated, randomized, open-label, clinical trial to determine the clinical effectiveness of oral sitafloxacin monotherapy in comparison to the conventional doxycycline–rifampicin combination regimen for acute uncomplicated human brucellosis.

Methods

In vitro antimicrobial susceptibility testing

A total of 30 clinical Brucella isolates collected from China were selected for antimicrobial susceptibility testing. The minimum inhibitory concentrations (MICs) of rifampicin, doxycycline, sitafloxacin, levofloxacin, and ciprofloxacin were determined using the broth microdilution method, with a customized susceptibility panel for aerobic Gram-negative bacilli (AST Panel for Aerobic Gram-Negative Bacilli, E-XW-Brucella) provided by Fosun Diagnostics (Changsha) Co. Ltd. The results were interpreted according to the Clinical and Laboratory Standards Institute (CLSI) guidelines for Brucella spp., and an MIC ≤1 μg/mL was considered to indicate susceptibility to doxycycline. As Brucella-specific thresholds were not available for the remaining antibiotics tested, interpretive criteria established for slow-growing bacteria (e.g., Haemophilus spp.) were applied.

Ethical considerations

For the multicenter, open-label, randomized, non-inferiority clinical trial, written informed consent was obtained from all patients or their legal representatives prior to study enrollment. The trial was conducted in compliance with the International Conference on Harmonization guidelines for good clinical practice (GCP) and the Declaration of Helsinki. The trial was approved by the Ethics Committee of Shandong University Qilu Hospital (approval NO. KYLL-202208-025-1) was prospectively registered at chictr. org.cn (NO. ChiCTR2300072703).

Patient selection

Patients were recruited from the clinics and Department of Infectious Disease at Qilu Hospital of Shandong University, Xinkang Hospital of Dalad Banner, and People's Hospital of Dalad Banner. Patients presenting with symptoms of acute brucellosis were consecutively screened for eligibility by a doctor according to pre-established criteria. Adult patients (aged ≥18 years) were eligible for enrollment if they presented with clinical manifestations consistent with brucellosis for less than three months and met at least one of the following diagnostic criteria: (1) isolation of any Brucella species from blood, bone marrow, or other body fluid, (2) standard tube agglutination test (SAT) titer ≥1:100, or (3) a positive IgG-ELISA (>12 U/ml) result. The exclusion criteria were (1) a history of hypersensitivity to sitafloxacin or other fluoroquinolones, doxycycline, or rifampicin; (2) pregnancy or breastfeeding; (3) a history of epilepsy or any other condition that results in predisposition to seizures; (4) diagnosis of myasthenia gravis; (5) a prolonged QT interval on electrocardiogram; (6) severe hepatic impairment (Child-Pugh grade C); (7) previous episode of tendinitis or tendon rupture following fluoroquinolone treatment; (8) severe renal insufficiency (creatinine clearance rate, <50 mL/min); (9) complicated brucellosis presentation (neurobrucellosis, spondylitis, and endocarditis); and (10) more than 1 week of treatment with anti-brucellosis agents in the past three months.

Amendments to the study protocol

After study initiation and before database lock, amendments to the study protocol were implemented following reassessment of study feasibility and recommendations from the investigators. The most important amendment was the change of the study design from a superiority trial to a noninferiority trial comparing sitafloxacin monotherapy with doxycycline–rifampicin for acute uncomplicated brucellosis.

Accordingly, the sample size estimation and statistical analysis plan were revised to align with the noninferiority framework. These amendments were finalized before completion of follow-up, database lock, and formal statistical analysis. No interim efficacy analysis was performed before the protocol amendment.

Randomized grouping of patients

Eligible patients were enrolled according to the final revised protocol and randomly assigned in a 1:1 ratio to either sitafloxacin monotherapy or doxycycline–rifampicin. The randomization sequence was generated using R software by an independent statistician, applying a permuted block design with variable block sizes. Allocation concealment was ensured using sequentially numbered, opaque, sealed envelopes, which were prepared by personnel not involved in patient enrollment. Envelopes were opened strictly in sequence only after patient inclusion, ensuring that treatment allocation could not be foreseen. Despite the open-label design, the following measures were implemented to minimize bias: two independent clinical experts, unaware of treatment allocation, reviewed all primary outcome events using predefined criteria. All investigators received training on uniform assessment procedures for adverse events and laboratory tests. Statistical analyses were performed by investigators blinded to group allocation.

Treatment regimens and evaluation of adherence and safety

Patients randomized to the sitafloxacin arm received oral sitafloxacin at a dosage of 100 mg once a day. Participants in the doxycycline–rifampicin combination therapy arm were prescribed oral doxycycline at a dosage of 100 mg every 12 h concurrently with weight-adjusted rifampicin (10 mg/kg daily). Treatment was administered for 6 weeks. To evaluate adherence to the prescribed regimen, a pill count methodology was implemented. During treatment initiation and subsequent clinical visits, the total number of doxycycline capsules, rifampicin capsules, and sitafloxacin tablets needed for 2 weeks of treatment were provided to each patient. Medication compliance was assessed by counting the number of pills remaining at each follow-up appointment.

The patients were given a diary for recording daily symptoms and were asked to record possible adverse effects (AEs) or complications. They were also asked to record whether they had completed the treatment. They were not allowed to take any additional NSAIDs during the trial, but were allowed to take ibuprofen if they needed additional pain relief. When the patients had completed the diary, they were asked to return it to the study site.

Assessment of clinical efficacy and safety

Prospective longitudinal follow-up was conducted in accordance with the trial protocol to evaluate both efficacy and safety outcomes. All patients were followed as outpatients for up to 12 months after therapy completion and were additionally assessed at any time if brucellosis-related symptoms recurred. The final follow-up visit was conducted on July 31, 2025. Clinical efficacy indicators included resolution of brucellosis-related symptoms and normalization of inflammatory markers, such as C-reactive protein, while safety was monitored through serial liver function tests and complete blood counts. These evaluations were performed at protocol-specified intervals: biweekly during the 6-week treatment phase (weeks 2, 4, and 6; allowable visit window ±3 days), monthly visits during the first 3 months post-treatment (months 1, 2, and 3; allowable visit window ±7 days), and quarterly visits through month 12 (months 6, 9, and 12; allowable visit window ±14 days). This tiered surveillance strategy, implemented with standardized data collection tools, ensured comprehensive documentation of treatment-emergent AEs, detection of delayed toxicities, and assessment of sustained therapeutic efficacy, with a particular focus on monitoring relapse and long-term clinical outcomes. To ensure data completeness, follow-up was actively conducted through scheduled outpatient visits supplemented by telephone contact when necessary; if a participant failed to attend a visit within the predefined window, repeated attempts were made to obtain outcome data remotely or to reschedule the assessment within an acceptable timeframe. When follow-up data could not be obtained despite these efforts, missing data were handled according to a prespecified strategy: in the intention-to-treat (ITT) analysis, patients with missing primary outcome data were conservatively classified as treatment failures, whereas in the per-protocol (PP) analysis, patients with missing primary outcome assessments were excluded.

Clinical outcomes

Clinical outcomes at the end-of-treatment and test-of-cure visits were classified by investigators as clinical response, relapse, or treatment failure. The primary outcome was the clinical response rate at the end of the 6-week treatment period, defined as the resolution of brucellosis-related symptoms without the need for additional antimicrobial therapy. The secondary outcomes included (1) clinical failure rate, defined as persistence or worsening of symptoms during treatment, need for treatment modification, or early discontinuation due to lack of efficacy and (2) relapse rate, defined as the reappearance of brucellosis-related symptoms and signs, confirmed by clinical and/or laboratory findings, during the post-treatment follow-up period in patients who initially achieved a clinical response.

AEs were summarized for events occurring from the first dose of study treatment to the final protocol follow-up visit. AEs were analyzed both by the proportion of patients experiencing at least one AE and by the frequency of individual AE types. Each AE symptom was recorded as a separate event, even if it occurred in the same patient.

Sample size estimation and statistical analyses

Sample size estimation was based on a non-inferiority trial design with a binary primary endpoint (clinical response rate). Assuming a clinical response rate of 87%^22,23 in the control group and an expected equal rate in the intervention group, with a non-inferiority margin of 12% (selected based on clinical and methodological considerations; see Supplementary Materials for detailed justification), a one-sided alpha of 0.05, and 80% power, a minimum of 98 patients per group was required. To account for an anticipated dropout rate of 10%, the final sample size was adjusted to 109 patients per group, resulting in a total of 218 patients. During the recruitment period, a total of 238 participants were ultimately enrolled (119 per group).

Efficacy endpoints were analyzed in the ITT and PP population. The ITT population comprised all randomized participants. Participants with missing primary outcome data due to loss to follow-up were conservatively classified as treatment failures in the ITT analysis. The PP population included only those who completed the entire 6-week therapeutic course without major protocol deviations. Major protocol deviations were defined as: (1) premature discontinuation of the assigned antimicrobial regimen before completion of the planned 6-week treatment course; (2) non-adherence to the randomized treatment allocation, including failure to initiate or switch from the assigned regimen; (3) use of concomitant non-protocol antimicrobial agents with known anti-Brucella activity during the treatment period; (4) poor treatment adherence, defined as receiving <80% of the prescribed doses.

For analysis of relapse, only patients who completed the treatment and post-treatment observation period were included. Participants who discontinued treatment early had already been classified as treatment failures and were, therefore, excluded from the relapse analysis.

The primary outcome was the clinical response rate at the end of treatment. Non-inferiority was assessed using the risk difference (sitafloxacin minus doxycycline–rifampicin). Treatment differences were presented as risk differences with 95% confidence intervals (CIs), calculated using the Miettinen–Nurminen method (score-based method for two independent proportions), implemented in R (version 4.5.1) using the PropCIs package (version 0.3-0). P values were calculated using the score test for two independent proportions without continuity correction. Non-inferiority was concluded if the lower bound of the two-sided 95% confidence interval exceeded the prespecified margin of −12%. Although the sample size was calculated based on a one-sided α of 0.05, the primary analysis was performed using two-sided 95% confidence intervals, corresponding to a one-sided α of 0.025, in line with conventional recommendations for noninferiority trials. Prior to participant recruitment, the study protocol was amended to refine the statistical design, including specification of the non-inferiority margin, adjustment of the sample size, and reclassification of treatment failure rate from a primary to a secondary endpoint. These changes were made before enrollment of the first participant and were not informed by study outcomes (see Supplementary Materials for details).

Differences between treatment groups were determined using a t-test for continuous variables and a Fisher's exact test or chi-square test for proportional data, with the significance level set at P < 0.05. Data were analyzed using SPSS statistical package, version 15.

Patient and public involvement

Patients and the public were not involved in the design, conduct, reporting, or dissemination of this study.

Role of the funding source

The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

Results

In vitro activity of sitafloxacin and other antibiotics against Brucella

For all 30 Brucella isolates, the MIC of rifampicin was consistently 0.5 μg/mL. Doxycycline demonstrated an MIC range of 0.06–0.12 μg/mL, with both MIC₅₀ and MIC₉₀ values being 0.12 μg/mL, resulting in a susceptibility rate of 100%. Sitafloxacin exhibited potent activity, with an MIC range of 0.03–0.12 μg/mL; the MIC₅₀ and MIC₉₀ were 0.06 μg/mL and 0.12 μg/mL, respectively. Levofloxacin displayed an MIC range of 0.25–0.5 μg/mL, with MIC₅₀ and MIC₉₀ values of 0.25 μg/mL and 0.5 μg/mL, respectively. Ciprofloxacin showed a similar MIC range of 0.25–0.5 μg/mL, with both MIC₅₀ and MIC₉₀ being 0.25 μg/mL (Table 1).

Table 1.

MIC values of antibiotics against 30 Brucella isolates.

Antibiotic	MIC range (μg/mL)	MIC50 (μg/mL)	MIC90 (μg/mL)	Threshold for susceptibility (μg/mL)
Rifampicin	0.5	0.5	0.5	≤1a
Doxycycline	0.06–0.12	0.12	0.12	≤1
Sitafloxacin	0.03–0.12	0.06	0.12	–
Levofloxacin	0.25–0.5	0.25	0.5	≤2a
Ciprofloxacin	0.25–0.5	0.25	0.25	≤1a

^aBased on CLSI thresholds for slow-growing bacteria (Haemophilus spp.).

Baseline characteristics of the study population

From August 1, 2023, to July 31, 2024, patients with acute uncomplicated human brucellosis who visited the three included hospitals were screened for eligibility. Of 431 patients assessed, 193 were excluded based on the predefined criteria listed in the methods section. A total of 238 eligible participants were randomized to the two treatment arms, with 119 patients in each arm, and all received at least one dose of the assigned study treatment (Fig. 1). The full analysis set (FAS) included 119 patients in the sitafloxacin group and 119 in the doxycycline plus rifampicin group. The cohort had a median age of 55 years (IQR, 47–62) and comprised predominantly male patients (69%, 164/238). The median disease duration was 28 days (IQR, 15–48). Key exposure histories included animal contact (68.1%, 162/238) and ingestion of contaminated products (22.7%, 54/238). Fever (81.5%, 194/238) and arthralgia (45.0%, 107/238) were the most common presenting symptoms. Brucella spp. were isolated from 113 patients for whom blood cultures were obtained. The baseline demographic and clinical characteristics were well matched between the two groups (Table 2).

Fig. 1.

Study flow diagram. In the intention-to-treat analysis, all randomized patients were included, and patients lost to follow-up were considered treatment failures.

Table 2.

Baseline characteristics of the intention-to-treat population.

Characteristics	Total (n = 238)	Doxycycline plus rifampicin (n = 119)	Sitafloxacin (n = 119)
Age (years)	55 (47–62)	54 (45–62)	55 (48–63)
Sex
Male	164 (69%)	83 (70%)	81 (68%)
Female	74 (31%)	36 (30%)	38 (32%)
Duration of symptoms (days)	28 (15–48)	30 (18–48)	26 (13–48)
Hypertension	31 (13.0%)	14 (11.8%)	17 (14.3%)
Diabetes mellitus	21 (8.8%)	13 (10.9%)	8 (6.7%)
Cardiovascular disease	19 (8.0%)	10 (8.4%)	9 (7.6%)
Cerebrovascular disease	11 (4.6%)	5 (4.2%)	6 (5.0%)
Malignancy	9 (3.8%)	3 (2.5%)	6 (5.0%)
Animal contact	162 (68.1%)	84 (70.6%)	78 (65.5%)
Dairy consumption	54 (22.7%)	23 (19.3%)	31 (26.1%)
Unknown	22 (9.2%)	12 (10.1%)	10 (8.4%)
Fever	194 (81.5%)	97 (81.5%)	97 (81.5%)
Arthralgia	107 (45.0%)	51 (42.9%)	56 (47.1%)
Fatigue	23 (9.7%)	12 (10.1%)	11 (9.2%)
Headache	14 (5.9%)	9 (7.6%)	5 (4.2%)
Weight loss	8 (3.4%)	6 (5.0%)	2 (1.7%)
Night sweat	7 (2.9%)	4 (3.4%)	3 (2.5%)
WBC count (109/L)	5.53 (4.27–7.04)	5.44 (4.24–6.83)	5.57 (4.32–7.40)
Hemoglobin level (g/L)	129 (118–141)	131 (118–142)	129 (119–141)
CRP level (mg/L)	19.14 (8.88–39.22)	18.20 (6.86–32.09)	21.58 (9.96–46.87)
BUN level (mmol/L)	4.6 (3.8–5.7)	4.5 (3.6–5.8)	4.6 (3.9–5.7)
Creatinine level (μmol/L)	61 (53–70)	60 (52–69)	61 (54–71)
AST level (U/L)	30 (20–52)	30 (19–45)	32 (21–52)
ALT level (U/L)	32 (20–54)	29 (18–50)	35 (24–64)
ALB level (g/L)	40 (37–43)	40 (37–44)	41 (38–43)
Culture-positive(n)	113	52	61

Data presented as n (%) or median (IQR p25–p75).

WBC, white blood cell; CRP, C-reactive protein; BUN, blood urea nitrogen; AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALB, albumin.

Treatment discontinuation rates

All randomized participants received at least one dose of the assigned study treatment. Treatment discontinuation was more frequent in the doxycycline–rifampicin group than in the sitafloxacin group (13.45% [16/119] vs. 7.56% [9/119]; p = 0.139). This difference was mainly attributable to a higher rate of AE-related discontinuation in the doxycycline–rifampicin group (6.72% [8/119]) than in the sitafloxacin group (2.52% [3/119]; risk difference = 4.2%; 95% CI = −10.5–1.3; p = 0.118), although the difference did not reach statistical significance. The occurrence of discontinuation due to perceived lack of efficacy was comparable between the two groups (4.20% [5/119] in the sitafloxacin group vs. 5.04% [6/119] in the doxycycline–rifampicin group; p = 0.758). Treatment was discontinued in one patient (0.84%) in the doxycycline–rifampicin group because it was discovered that the patient met an exclusion criterion after randomization into the treatment group.

Efficacy outcomes

In the ITT population, which included all randomized participants, the clinical response rate at the end of therapy was higher in the sitafloxacin group (89.9%, 107/119) than in the doxycycline–rifampicin group (83.2%, 99/119). The risk difference was 6.7% (95% CI: −2.0%–15.7%, p = 0.135), as calculated using Miettinen–Nurminen method. The lower bound of the 95% CI (−2.0%) exceeded the prespecified non-inferiority margin of −12%, demonstrating non-inferiority of sitafloxacin to doxycycline-rifampin.

The PP analysis, which was limited to participants who completed the full 6-week course without major protocol deviations, yielded consistent results: 97.3% clinical response rate (107/110) for sitafloxacin vs. 96.1% (99/103) for doxycycline–rifampicin, with a risk difference of 1.2% (95% CI: −4.4%–7.2%, p = 0.641) (Table 3). This indicates that the treatment regimens had similar efficacy in adherent populations, which also met the criterion for non-inferiority.

Table 3.

Comparison of clinical response, treatment failure, and relapse rates between the sitafloxacin and doxycycline–rifampicin treatment groups.

Outcome	Sitafloxacin (n = 119)	Doxycycline + rifampicin (n = 119)	Risk difference (95% CI)	P value
Clinical response rate in the ITT group, n (%)	107 (89.9%)	99 (83.2%)	6.7% (−2.0 to15.7)	0.135
Clinical response rate in the PP group, n (%)	107/110 (97.3%)	99/103 (96.1%)	1.2% (−4.4 to 7.2)	0.641
Relapse rate in the PP group, n (%)	3/110 (2.7%)	2/103 (1.9%)	0.8% (−4.4 to 6.0)	0.673

Data are presented as number of patients (n) and percentage (%) of the total group population (n = 119 per arm). P values were calculated using chi-square or Fisher's exact tests, as appropriate. Non-inferiority was assessed based on the two-sided 95% confidence interval for the risk difference. The ITT clinical failure rate is the complement of the ITT clinical response rate and was not separately analyzed to avoid duplication of statistical testing.

As predefined by the trial protocol, patients who discontinued treatment prematurely regardless of the reason were conservatively classified as treatment failures in the primary ITT efficacy analysis. Accordingly, the overall clinical failure rate in the ITT population was 10.1% (12/119) in the sitafloxacin group and 16.8% (20/119) in the doxycycline–rifampicin group, yielding a risk difference of −6.7% (95% CI: −15.7%–2.0%; p = 0.135) in favor of sitafloxacin. These cases of failure included both cases of early treatment discontinuation and cases with documented lack of clinical improvement at the end of therapy.

Among patients who achieved an initial clinical response and completed the 6-week treatment course, 3 patients (2.7%) in the sitafloxacin group and 2 patients (1.9%) in the doxycycline–rifampicin group experienced symptom recurrence that was confirmed by clinical and laboratory evaluation. Detailed characteristics of patients with relapse are shown in Table 4. The risk difference for symptom recurrence between the sitafloxacin and doxycycline-rifampicin groups was 0.8% (95% CI: −4.3%–6.0%), with no statistically significant difference observed (p = 0.673).

Table 4.

Characteristics of patients with relapse (n = 5).

Case	Sex, Age (y)	Initial disease				Relapse
		Duration of symptoms (days)	Blood culture	CRP (mg/L)	Comorbidity	Time to Relapse (months)	Clinical findings	Blood culture result
Sitafloxacin
1	F, 56	30	Not performed	9.05	Diabetes mellitus	3	Fever Neck pain Cervical spondylitis	N
2	M, 34	10	P	30.51	N	2	Fever Orchitis	N
3	M, 49	6	P	48	N	2	Fever Lower back pain Lumbar spondylitis	N
Doxycycline–Rifampicin
1	M,28	30	N	32.8	N	2	Fever Orchitis	N
2	M,26	30	Not performed	10.5	N	2	Fever	N

M, Male; F, Female; N, Negative; P, Positive.

Treatment safety

In the sitafloxacin group, 10.9% (13/119) of patients reported at least one treatment-related AE, as compared to 19.3% (23/119) in the rifampicin–doxycycline group (p = 0.07). The most common AEs in the sitafloxacin group included gastrointestinal symptoms (nausea, diarrhea, and vomiting), rash, pruritus, and elevations in liver enzymes or bilirubin. The risk difference for treatment-related AEs numerically favored sitafloxacin, with an observed reduction of −8.4% (95% CI: −17.7% to 0.7%), although the difference did not reach statistical significance. A full list of AEs by category and severity is presented in Table 5.

Table 5.

Comparison of treatment-related adverse event profiles between sitafloxacin and doxycycline–rifampicin therapy in the intention-to-treat population.

Adverse events	Sitafloxacin (n = 119)	Doxycycline–rifampicin (n = 119)	P value
Patients with ≥1 AE	13 (10.9%)	23 (19.3%)	0.07
Nausea	5 (4.2%)	10 (8.4%)	0.18
Diarrhea	3 (2.5%)	2 (1.7%)	1.00a
Vomiting	0	4 (3.4%)	0.12a
Rash	2 (1.7%)	0	0.50a
Pruritus	2 (1.7%)	0	0.50a
Elevated liver enzymes	1 (0.8%)	4 (3.4%)	0.37a
Hyperbilirubinemia	0	4 (3.4%)	0.12a

Data are presented as number of patients (n) and percentage (%) of the total population (n = 119 per arm). Percentages reflect the proportion of patients experiencing each AE type, with multiple events per patient counted individually.

^aP-values calculated using Fisher's exact test.

Discussion

The present study demonstrates the clinical efficacy and safety of monotherapy with sitafloxacin in cases of human brucellosis. In addition, we have also confirmed previous in vitro findings on the susceptibility of Brucella spp. to this fluoroquinolone. Importantly, this is the first randomized controlled trial to evaluate the potential efficacy of sitafloxacin treatment for human brucellosis infection.

In our study, sitafloxacin, the most active of the other two tested fluoroquinolones (levofloxacin and ciprofloxacin), showed excellent in vitro activity against all strains of Brucella melitensis. Across the 30 Brucella clinical isolates obtained from China, the MIC₅₀ and MIC₉₀ of sitafloxacin were 0.06 μg/mL and 0.12 μg/mL, respectively. These values are consistent with those reported by Martín et al.¹⁷ Sitafloxacin has been found to exhibit high activity against most clinical isolates in hospitals across China,²⁴ as well as specific bacterial strains. For example, Choi tested the anti-bacterial effect of sitafloxacin against 121 H. pylori strains and found that none of them were resistant to this fluoroquinolone (MIC >1, n = 0).²⁵ Further, a study of the MICs of sitafloxacin against Mycobacterium abscessus isolates from 86 patients showed that 51 isolates (59.3%) had MIC values less than 1 μg/mL.²⁶ All these findings imply that sitafloxacin exhibits activity against both intracellular and extracellular bacteria and support the observed in vitro effects on Brucella in this study. Altogether, they suggest that sitafloxacin would be a good candidate for the treatment of human brucellosis.

The results of our clinical trial are equally encouraging. The results of our clinical trial support the in vitro findings and provide robust evidence for the therapeutic potential of sitafloxacin monotherapy for brucellosis. In the ITT population, sitafloxacin demonstrated non-inferior efficacy compared to the conventional doxycycline–rifampicin combination, with a clinical success rate of 89.9% (107/119) for the former and 83.2% (99/119) for the latter. Similar cure rates have been reported for doxycycline–rifampicin in the literature (81% and 83.9%).^23,27 In the present study, sitafloxacin showed a numerically more favorable safety profile, with a lower incidence of treatment-related AEs compared to doxycycline–rifampicin (10.9% vs. 19.3%; p = 0.07), although the difference did not reach statistical significance. Further, the recurrence rate was not significantly different between the two treatment arms. These findings suggest that a single-agent regimen may be a potential alternative, with a trend toward fewer multidrug-related toxicities. It should be noted that the recurrence rate with doxycycline–rifampicin treatment was 1.9%, which is considerably lower than the recurrence rates of 16% and 15.3% reported in the literature.^22,23 This is probably because the cases in the present were all non-complicated cases in the acute phase. Whereas the other two studies also included complicated and chronic cases.

The PP analysis, limited to participants who completed the full 6-week course without major protocol deviations, yielded consistent results: 97.3% for sitafloxacin vs. 96.1% for doxycycline–rifampicin, with a risk difference of 1.2%. We also found that treatment with sitafloxacin was associated with a lower rate of withdrawal, although this difference did not reach statistical significance. Overall, the relapse rate, fever-free interval before relapse, and drug tolerance were similar for both treatments. These basic and clinical findings demonstrate that the effect of sitafloxacin in the treatment of Brucella infection is not inferior to that of conventional combination therapy. While the cost of conventional treatment is lower, the single-agent regimen might be more convenient for patients because the number of tablets ingested per day is reduced by nearly one-third. The convenience of single-agent application may translate into better adherence and overall success rates.

As with most fluoroquinolones, the clinical use of sitafloxacin requires careful consideration of patient benefits and potentially life-threatening risks. Common AEs of quinolones that should be monitored in all patients include, but are not limited to, gastrointestinal (gastritis and hepatotoxicity), neurologic (altered mental status and peripheral neuropathy), cardiovascular (prolonged QT interval leading to torsades de pointes, a fatal arrhythmia), tendinopathy, dysglycemia, and phototoxicity.²⁸ In particular, the common side effects of sitafloxacin are more localized to the gastrointestinal tract; accordingly, nausea and diarrhea were observed in the present study (in five and three cases, respectively). Further, rash and pruritis were also observed in two patients each, as was elevated liver enzyme levels (in one patient). Other studies have reported dose-dependent mild phototoxic effects in white patients that were not observed in Asian patients, as well as dizziness (the most common clinical AE in a group administered at 100 mg qd).28, 29, 30 While the present patients did not exhibit several of the previously reported AEs. Extensive safety surveillance data are required for sitafloxacin before safety can be definitively established.

Limitations

This study has several limitations. First, its open-label design is subject to potential performance and detection bias, particularly in terms of the subjective endpoints and ancillary treatments. However, this risk was mitigated through blinded adjudication of key clinical outcomes and the use of objective primary efficacy measures. Second, relapse analysis was restricted to patients who completed the full treatment course. Excluding those who discontinued early and were not followed longitudinally may have led to an underestimation of the true relapse rates. Third, the findings are limited to patients with acute uncomplicated brucellosis and may not be generalizable to those with complicated disease forms such as neurobrucellosis and spondylitis. Despite careful baseline evaluation and prospective follow-up to identify focal complications, occult or early-stage focal infections cannot be completely excluded; therefore, sitafloxacin monotherapy should be applied with caution when focal involvement is suspected. Fourth, the trial investigated the effects of a fixed 6-week regimen, and it is not clear whether shorter sitafloxacin courses could offer similar efficacy with enhanced tolerability. Finally, the absence of a cost-effectiveness evaluation limits the applicability of these results in resource-limited settings where brucellosis remains endemic. Future studies in more ethnically diverse populations are warranted to validate the external generalizability of our findings.

Conclusion

In conclusion, sitafloxacin monotherapy demonstrated non-inferior efficacy and favorable tolerability profile compared with the standard doxycycline–rifampicin regimen in the treatment of acute uncomplicated brucellosis. These findings indicate that sitafloxacin may serve as a potential alternative treatment option, particularly in patients at higher risk for adverse events or poor adherence. However, further studies are warranted to confirm its effectiveness in complicated disease forms and with shorter treatment durations, as well as to assess cost-effectiveness in endemic and resource-constrained settings.

Contributors

CMQ, NNX, GW, and HJ were involved in the design, establishment, and day-to-day management and implementation of the trial. GW, and HJ obtained funding for the trial. CMQ, MHX, XG, ZRL, LJG, and GW enrolled participants in the trial. QQX was responsible for in vitro antimicrobial susceptibility testing. CMQ, QQX, LHZ, and NNX were in charge of data curation and accessed and verified the data and involved in the statistical analyses. CMQ, QQX, and NNX wrote the original draft of the manuscript, which was reviewed and edited by GW, and HJ. All authors contributed to refinement of and approved this manuscript. All authors had full access to all the data and accept the responsibility to submit for publication.

Data sharing statement

The data utilized in this study are available from the corresponding author on reasonable request.

Protocol and statistical analysis plan

The trial protocol and statistical analysis plan are available from the corresponding author upon reasonable request.

Declaration of interests

The authors have no conflicting interests to declare.