Treatments for bacterial vaginosis currently recommended by the Centers for Disease Control and Prevention (CDC) include multidose oral and vaginal formulations of metronidazole and clindamycin.1 Alternative regimens include oral formulations of tinidazole or clindamycin and intravaginal formulations of clindamycin and metronidazole.2,3 Currently, there are no U.S. Food and Drug Administration (FDA)–approved single-dose oral therapy options for the treatment of bacterial vaginosis.
Secnidazole, a 5-nitroimidazole with a longer half-life than metronidazole (approximately 17 compared with approximately 8 hours), has been used for the treatment of bacterial vaginosis, trichomoniasis, and other indications in Europe and Asia4 but is not available in the United States. A single dose of 2 g secnidazole was compared with a standard 7-day regimen of metronidazole in a European study of women with Nugent scores 7 or greater and found similar resolution of bacterial vaginosis between the treatment groups.5 In vitro studies demonstrated antimicrobial properties of secnidazole against most of the bacterial species implicated in bacterial vaginosis. Several published clinical studies have provided both clinical and microbiologic evidence of activity of secnidazole in the treatment of bacterial vaginosis.5–7 Secnidazole, like other 5-nitroimidazoles, has limited activity against lactobacilli, the beneficial microbes important in establishing vaginal health after treatment.
This three-arm, placebo-controlled, phase 2, dose-ranging study of secnidazole for bacterial vaginosis treatment was conducted in accordance with FDA guidance. The objective was to evaluate the clinical, microbiologic, and therapeutic efficacy and safety of a single oral dose of 1 or 2 g secnidazole in women diagnosed with bacterial vaginosis, based on Amsel criteria and Nugent scores 4 or greater.
MATERIALS AND METHODS
The study received institutional review board approval (Schulman Associates institutional review board, Cincinnati, Ohio; Western institutional review board, Puyallup and Olympia, Washington; Wayne State University Human Investigation Committee, Detroit, Michigan). Patients provided written informed consent before any study procedures were performed. Patients determined to be eligible at baseline (day 1) were randomized one to one to one using a centralized permuted block randomization process through the data management center to one of the following: one sealed bottle of granules containing 1 or 2 g secnidazole or matching placebo granules. The sealed bottles contained the randomization assignment, which was generated by the study biostatistician before the start of the study. All ingredients of the granules were the same with the exception of the addition of a bittering agent to the placebo granules to mimic the potential bitter taste of secnidazole granules. All treatment was administered orally as a single dose in approximately 4 ounces of applesauce under fasted conditions (ie, no food 2 hours before through 1 hour after dosing). After baseline assessments, patients self-administered treatments while at the clinic. Randomization was stratified by the number of self-reported episodes of bacterial vaginosis, including the current episode, in the past 12 months (three or fewer compared with four or more) as reported during the medical history. Patients were contacted by telephone once between days 8 and 10 to inquire about possible adverse events and persistence of symptoms. Patients completed a test of cure visit, comprising assessments pertaining to primary and secondary efficacy endpoints and the safety endpoint between days 21 and 30 posttreatment. Patients were free to withdraw at any time for any reason. Women who requested alternative treatment at any time during the study were offered standard of care bacterial vaginosis treatment at no cost. If a patient withdrew before the test of cure visit, an end of study visit, comprising a test of cure assessments, was performed at the time of withdrawal. This study was conducted in accordance with the ethical principles of Good Clinical Practice.8
Patients with a clinical diagnosis of bacterial vaginosis were recruited from 24 ambulatory gynecology clinics in the United States, and by design, no single site contributed more than 16% to the study population. The study population included nonpregnant women who were 18 years of age or older, in general good health, had agreed to abstain from sexual activity and use of intravaginal products for 1 week after treatment, and met the four Amsel criteria for bacterial vaginosis (discharge; pH 4.7 or greater; 20% or greater clue cells; positive whiff test). Women were excluded if they were pregnant, bleeding at baseline, were allergic to metronidazole, unable or unwilling to abstain from alcohol for 3 days after treatment, receiving concomitant antimicrobial therapy (topical or oral), or received a recent course (within previous 14 days) of antimicrobial or antifungal therapy. Baseline vaginal sample slides for Nugent scoring and laboratory tests confirming sexually transmitted infections (STIs) were evaluated centrally; thus, results were unavailable at the time of randomization. The study protocol prespecified that patients determined to have Nugent scores less than 4 or who tested positive for a STI after baseline assessment were to be considered screening failures. These patients underwent an end of study visit and were then discontinued from the study. Accordingly, these patients received the randomized treatment and remained in the study until Nugent scores and baseline laboratory tests became available to the investigator. Baseline Nugent scores ultimately were used as a criterion for inclusion in the modified intent-to-treat population, the primary population for efficacy analyses (Fig. 1).
The primary efficacy endpoint was clinical cure, based on the 1998 FDA guidance regarding evaluation of treatment for bacterial vaginosis: 1) normal vaginal discharge, 2) negative 10% potassium hydroxide whiff test, and 3) clue cells less than 20% of total epithelial cells on microscopic examination of the vaginal wet mount using saline at the test of cure visit. Secondary efficacy endpoints, also evaluated at this visit, were Nugent score, with a score of 0–3 considered Lactobacillus-dominant (ie, microbiologic cure) and a score of 4 or greater considered abnormal, and therapeutic cure, defined as meeting the criteria for both clinical and microbiologic cure.
Safety evaluations were based on the incidence, intensity, and type of adverse events and changes in patients' physical examination findings, vital signs, and clinical laboratory results, but the study was not statistically powered to detect differences in safety between groups.
Assuming a clinical cure rate of 40% in the active groups and 15% in the placebo group, a sample size of 52 patients per group was expected to provide approximately 80% power to detect a statistically significant difference between either secnidazole group and placebo at a .05 level of significance (two-sided) using a Cochran-Mantel-Haenszel test. To ensure at least 52 patients met the modified intent-to-treat criteria, between 71 and 72 patients were randomized to each treatment group for a total of 215 patients randomized.
No interim analyses of efficacy were performed and sponsors and investigators remained blinded until the database was locked. The efficacy analyses were performed on the modified intent-to-treat population, defined as all randomized patients who met all study selection criteria. Clinical cure at the test of cure visit was compared between each active secnidazole dose and placebo using a Cochran-Mantel-Haenszel statistical test adjusted for stratification by bacterial vaginosis recurrence. The primary comparison was between 2 g secnidazole and placebo; comparison between 1 g secnidazole and placebo was a secondary analysis. The 2- and 1-g secnidazole doses were not compared statistically. Because there was only one primary comparison, no adjustment for multiple comparisons was indicated. For each secnidazole dose, an exact 95% binomial confidence interval (CI) was calculated for the clinical cure rate. All tests were performed at a .05 level of significance (two-sided). Secondary efficacy endpoints were analyzed similarly.
In a prespecified analysis, primary and secondary endpoints were also analyzed by bacterial vaginosis strata and baseline Nugent score (4–6, 7–10). Patients with any missing data required for assessment of primary and secondary efficacy endpoints at either the test of cure or end of study visit as a result of treatment failure or early discontinuation were imputed as nonresponders for clinical and therapeutic outcomes and as abnormal for Nugent score.
Safety evaluations were based on the incidence, intensity, and type of treatment-emergent adverse events and changes in the patient's physical examination findings, vital signs, and clinical safety laboratory results. Safety variables were tabulated and presented for all patients in the intent-to-treat population, because they all received the study drug. A treatment-emergent adverse event was defined as any adverse event that occurred after administration of the study drug dose through the end of the study visit. Any event that was considered by the investigator to be study drug–related, regardless of onset time, or any event that was present at baseline but worsened in intensity or was subsequently was considered treatment-related by the investigator. Investigators were asked to report all adverse events, whether mild, moderate, or severe. Adverse events were recorded verbatim and then coded using preferred Medical Dictionary for Regulatory Activities terminology.
A total of 215 women were enrolled at 24 research centers in the United States between May 28 and September 5, 2014. Reasons for exclusion included baseline STIs (n=13 women), Nugent score less than 4 (n=14), and clinically significant baseline laboratory abnormality that led to a diagnosis of chronic myeloid leukemia (n=1; Fig. 1), leaving 188 women in the modified intent-to-treat population with 62–64 women per arm. The demographic and baseline characteristics of the study population are summarized in Table 1 and Appendix 1 (available online at http://links.lww.com/AOG/A973). The median age of the study population was 33 years (range 19–54 years); more than half of the participants (54%) described themselves as black or African American, and 39% identified as white. A higher proportion of white patients were enrolled in the 2-g secnidazole group (51.6%) compared with the 1-g and placebo groups (28.1% and 38.7%, respectively), although these differences were not statistically significantly different. Other demographics and baseline characteristics were similar across the three treatment arms.
In the intent-to-treat population, the clinical cure rate was 65.3% for the 2-g group, 49.3% for the 1-g group, and 19.4% for the placebo group. However, as prespecified in the protocol, the modified intent-to-treat group was designated the primary analysis population, because two different criteria for exclusion—baseline Nugent scores and results from baseline laboratory testing—were only available after the patients were randomized and treated. Clinical, microbiologic, and therapeutic cure rates are summarized for the modified intent-to-treat population in Table 2. Eight women were lost to follow-up during the course of the study and, per the prespecified protocol, were counted as nonresponders.
The clinical cure rate was significantly higher for the 2-g (68%) and 1-g (52%) doses of secnidazole compared with placebo (18%) (P<.001 for both comparisons). A similar trend was observed with microbiologic cure, which was 40% for the 2-g dose (P<.001 compared with placebo) and 23% for the 1-g dose (P=.007 compared with placebo). As shown in Table 2, the therapeutic cure rate was 40%, 22%, and 7% for the 2-g secnidazole, the 1-g, and the placebo groups, respectively. A greater treatment effect was observed with 2 g secnidazole compared with placebo on analysis of clinical, microbiologic, and therapeutic cure, regardless of the number of prior episodes of bacterial vaginosis (three or fewer or four or more in the past 12 months) in the modified intent-to-treat population. The study was not powered to determine statistical differences in a three-way stratified comparison. However, with respect to the nominal α=.05 level of significance, all three outcomes were favorable in the 2-g treatment arm in the subpopulation of patients with four or more episodes (P<.001 for all assessments) and for the clinical cure assessment for the 1-g treatment arm (P=.023; Table 3). Analyses of the primary and secondary endpoints in the per-protocol population (1-g group, n=55; 2-g group, n=47; placebo group, n=50), which included patients in the modified intent-to-treat population who had no major protocol violations, also showed results similar to those in the modified intent-to-treat population (clinical cure: 1 g, 52.7%, P<.001; 2 g, 72.3%, P<.001; placebo, 20%).
The safety and tolerability of secnidazole administered in the 2- or 1-g granule form was evaluated in all 215 randomized patients who were treated with a single dose of the study drug (Table 4). Treatment-emergent adverse events were reported in 19% (14/72), 13% (9/71), and 10% (7/72) in the 2-g secnidazole, 1-g secnidazole, and placebo groups, respectively. Infections were the most common events in all treatment groups, but were similar in incidence across the three treatment groups (Table 4). Vaginal yeast infections were infrequent, occurring in seven women, five of whom received secnidazole. All treatment-emergent adverse events were mild or moderate in intensity. No serious adverse events were reported. No secnidazole dose relationship or differences from placebo were observed in physical examination, clinical laboratory, or vital sign parameters. Furthermore, no notable differences were observed in the mean, median, minimum, or maximum changes from baseline in any laboratory parameter.
This study evaluated the efficacy and safety of two doses of secnidazole. Both dose levels of secnidazole were superior to placebo, although the study was not powered to detect differences between the two treatment groups. This study had a balanced inclusion of women having recurrent bacterial vaginosis over the three treatment arms. Although cure rates were lower overall among women having recurrent bacterial vaginosis, the 2-g secnidazole treatment was statistically superior to placebo, even among those women having recurrent infections. A strength of the study was the recruitment of women from across the United States and substantial representation by African American women.
Since this study was conducted, the FDA has issued new draft guidance regarding drug development for bacterial vaginosis. Interpreting the results of this study requires an understanding of this context, in which the FDA recommends the inclusion of a concurrent placebo arm as part of the design and that clinical cure being defined as the resolution of three clinical signs (discharge, amine odor, clue cells) of bacterial vaginosis.9 Using this definition, cure rates for oral 500 mg metronidazole twice daily for 7 days have been reported to be 58%5; the 7-day clindamycin cream treatment has provided a similar response.2 In a multicenter, placebo-controlled randomized trial of oral tinidazole given at either 2 g daily for 2 days or 1 g for 5 days, resolution of bacterial vaginosis as defined by absence of three or more Amsel criteria was reported in 46% of women who received the 2-day regimen and in 64% of those who received the 5-day regimen.10
Two single-dose intravaginal treatments have been FDA-approved for treatment of bacterial vaginosis. Single-dose clindamycin cream has been reported to provide clinical and microbiologic cure for 64% and 57% of women, respectively, although only 47% of enrolled women were evaluable in this study.2 A single-dose 1.3% metronidazole gel has a reported clinical cure rate of 37% and a microbiologic cure of 18% at the 21-day visit.3 In the present study, the single-dose oral treatment with secnidazole granules provided a clinical cure rate of 68% and a microbiologic cure rate of 40%, which appears to be similar to the efficacy of the current CDC-recommended regimens and similar or superior to the published data for single-dose clindamycin2 and metronidazole3 intravaginal treatments. Treatment with a single dose of oral secnidazole yielded similar cure rates when compared with 7 days of oral metronidazole in a European study.5 A consistent trend across studies is the lower microbiologic cure rate than the clinical rate, suggesting the presence of microbes persist after resolution of clinical symptoms, and lactobacilli species have not yet recovered to levels reflective of a healthy vaginal environment. An improved understanding of the tipping point, the point at which the imbalance of vaginal microbiota leads to signs and symptoms of bacterial vaginosis, is warranted to further understand whether differences in microbiologic cure affect risk of recurrence.
Side effects of oral metronidazole include nausea, a metallic taste, headaches, and gastrointestinal distress. In this study, the incidence of adverse events was low, with only 1 of 144 women randomized to secnidazole having nausea and only one woman reporting a metallic taste. The low incidence of adverse events is consistent with clinical experience with secnidazole as treatment outside the United States where it is approved for treatment of parasitic diseases and bacterial vaginosis with a well-established safety profile.5–7,11,12
The 2015 CDC Sexually Transmitted Disease Treatment Guidelines recommends treatment of bacterial vaginosis for all women having symptoms.1 Although the primary objective of treatment is resolution of symptoms, other potential benefits of bacterial vaginosis treatment include reduction in the risk for acquiring Chlamydia trachomatis, Neisseria gonorrhoeae,13 Trichomonas vaginalis,14 human immunodeficiency virus,15 and herpes simplex type 2.16 Recent studies have evaluated the effect of bacterial vaginosis treatment on acquisition of bacterial STIs with mixed results.17,18 Schwebke18 evaluated home screening for bacterial vaginosis to identify and treat asymptomatic cases to reduce acquisition of chlamydial and gonococcal infections, but found no difference in incident STIs. Results from a study of periodic presumptive treatment of bacterial vaginosis and yeast vaginitis demonstrated a reduced incidence of C trachomatis, N gonorrhoeae, and Mycoplasma genitalium.17 Additional research is needed to confirm the benefits of periodic treatment of bacterial vaginosis as a strategy to reduce incident STIs.
Results from this study support the further development of secnidazole for the treatment of women with bacterial vaginosis. Additional studies comparing single-dose 2-g secnidazole oral granules with an active comparator such as metronidazole may be warranted to better understand how a single-dose treatment may compare with a multidose regimen.
1. Workowski KA, Bolan GA; Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines, 2015. MMWR Recomm Rep 2015;64:1–137.
2. Faro S, Skokos CK; Clindesse Investigators Group. The efficacy and safety of a single dose of Clindesse vaginal cream versus a seven-dose regimen of Cleocin vaginal cream in patients with bacterial vaginosis. Infect Dis Obstet Gynecol 2005;13:155–60.
3. Schwebke JR, Marrazzo J, Beelen AP, Sobel JD. A phase 3, multicenter, randomized, double-blind, vehicle-controlled study evaluating the safety and efficacy of metronidazole vaginal gel 1.3% in the treatment of bacterial vaginosis. Sex Transm Dis 2015;42:376–81.
4. Videau D, Niel G, Siboulet A, Catalan F. Secnidazole. A 5-nitroimidazole derivative with a long half-life. Br J Vener Dis 1978;54:77–80.
5. Bohbot JM, Vicaut E, Fagnen D, Brauman M. Treatment of bacterial vaginosis: a multicenter, double-blind, double-dummy, randomised phase III study comparing secnidazole and metronidazole. Infect Dis Obstet Gynecol 2010 [epub ahead of print].
6. Donders G. Diagnosis and management of bacterial vaginosis and other types of abnormal vaginal bacterial flora: a review. Obstet Gynecol Surv 2010;65:462–73.
7. Núñez JT, Gómez G. Low-dose secnidazole in the treatment of bacterial vaginosis. Int J Gynaecol Obstet 2005;88:281–5.
8. World Medical Association. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA 2013;310:2191–4.
9. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER). Bacterial vaginosis: developing drugs for treatment. Guidance for industry. Silver Spring (MD): U.S. Food and Drug Administration; 2016.
10. Livengood CH III, Ferris DG, Wiesenfeld HC, Hillier SL, Soper DE, Nyirjesy P, et al. Effectiveness of two tinidazole regimens in treatment of bacterial vaginosis: a randomized controlled trial. Obstet Gynecol 2007;110:302–9.
11. Gillis JC, Wiseman LR. Secnidazole. A review of its antimicrobial activity, pharmacokinetic properties and therapeutic use in the management of protozoal infections and bacterial vaginosis. Drugs 1996;51:621–38.
12. Moraes ME, Cunha GH, Bezerra MM, Fechine FV, Pontes AV, Andrade WS, et al. Efficacy of the Mentha crispa in the treatment of women with Trichomonas vaginalis
infection. Arch Gynecol Obstet 2012;286:125–30.
13. Wiesenfeld HC, Hillier SL, Krohn MA, Landers DV, Sweet RL. Bacterial vaginosis is a strong predictor of Neisseria gonorrhoeae and Chlamydia trachomatis infection. Clin Infect Dis 2003;36:663–8.
14. Brotman RM. Vaginal microbiome and sexually transmitted infections: an epidemiologic perspective. J Clin Invest 2011;121:4610–7.
15. Atashili J, Poole C, Ndumbe PM, Adimora AA, Smith JS. Bacterial vaginosis and HIV acquisition: a meta-analysis of published studies. AIDS 2008;22:1493–501.
16. Cherpes TL, Meyn LA, Krohn MA, Lurie JG, Hillier SL. Association between acquisition of herpes simplex virus type 2 in women and bacterial vaginosis. Clin Infect Dis 2003;37:319–25.
17. Balkus JE, Manhart LE, Lee J, Anzala O, Kimani J, Schwebke J, et al. Periodic presumptive treatment for vaginal infections may reduce the incidence of sexually transmitted bacterial infections. J Infect Dis 2016;213:1932–7.
18. Schwebke JR, Lee JY, Lensing S, Philip SS, Wiesenfeld HC, Seña AC, et al. Home screening for bacterial vaginosis to prevent sexually transmitted diseases. Clin Infect Dis 2016;62:531–6.