Tioconazole

This draft guidance, once finalized, will represent the Food and Drug Administration's (FDA's) current thinking on this topic. It does not create or confer any rights for or on any person and does not operate to bind FDA or the public. You can use an alternative approach if the approach satisfies the requirements of the applicable statutes and regulations. If you want to discuss an alternative approach, contact the Office of Generic Drugs.

1. The Office of Generic Drugs (OGD) recommends a clinical endpoint bioequivalence study comparing the Tioconazole Vaginal Ointment, 6.5% test product versus the reference listed drug (RLD) and placebo, each administered by inserting one full applicator of ointment into the vagina once (study Day 1), with the primary endpoint evaluation at the test-of-cure visit conducted on study Day 21-30 (i.e., at the visit occurring between 20-29 days after administration of the last dose of vaginal ointment).
2. It is the sponsor's responsibility to enroll sufficient patients for the study to demonstrate bioequivalence between the products.
3. Assignment of the test product, RLD, and placebo should be randomized. The method of randomization should be described in the protocol. It is recommended that an independent third party generate and hold the randomization code throughout the conduct of the study in order to minimize bias. The sponsor may generate the randomization code if not involved in the packaging and labeling of the study medication. A sealed copy of the randomization scheme should be retained at the study site and should be available to FDA investigators at the time of site inspection to allow for verification of the treatment identity of each subject.

4.
A detailed description of the blinding procedure is to be provided in the protocol. The packaging of the test, reference and placebo products should be similar in appearance to make differences in g. History of allergy or sensitivity to tioconazole, related compounds or any component of the formulation.
8. Vaginal discharge should not be an inclusion criterion or included in the evaluation of treatment as this sign cannot be consistently correlated with the presence or absence of VVC.

Recommended Mar 2011
9. Positive vaginal fungal culture at baseline should not be an inclusion criterion due to the time lag between obtaining the culture specimen and receiving the culture results. However, a vaginal fungal culture must be obtained at baseline. Testing should be performed to identify the isolates to the species level (e.g., Candida albicans, Candida tropicalis, or Candida glabrata). Only patients with a pretreatment, baseline vaginal fungal culture that is positive for a Candida species should be included in the per protocol (PP) and modified intent to treat (mITT) populations for the primary endpoint analysis.
10. Provide a listing of the prescription and over-the-counter drug products that are prohibited during the study, such as: a. Topical products applied to the vulva or vagina (e.g., antibiotic, antifungal, antitrichomonal, corticosteroid or anti-inflammatory topical products). b. Vaginal products other than study product (e.g., vaginal estrogen, vaginal progesterone, douches, spermicides, condoms, tampons, diaphragms, contraceptive cream, contraceptive foam, or contraceptive film). c. Oral antibiotics, antifungals, or antitrichomonals. d. Oral or injectable corticosteroid or immunosuppressive.
11. The recommended primary endpoint of the study is the proportion of patients with therapeutic cure, defined as both mycological cure and clinical cure, at the test-of-cure visit conducted on study Days 21-30. Mycological cure is defined as a negative KOH wet mount test of vaginal secretions AND a negative vaginal fungal culture for Candida species. Clinical cure is defined as ALL of the following: a. Any vulvovaginal sign or symptom with a baseline score of 1 or 2 (on a 4-point scale as provided in Comment #12) has a score of 0 (absent) at the test-of-cure visit on study Day 21 30. b. Any vulvovaginal sign or symptom with a baseline score of 3 (on a 4-point scale as provided in Comment #12) has a score of 0 (absent) or 1 (mild) at the test-of-cure visit on study Day 21-30. c. Any new sign or symptom observed at the test-of-cure visit is determined by the investigator not to be related to VVC. d. The subject did not use any topical drug therapy (such as topical analgesics or corticosteroid products) other than the study product for the treatment of vulvovaginal irritation and/or pruritus.
12. Each of the following six vulvovaginal signs and symptoms should be individually scored using an accepted scale and then added together to determine the composite vulvovaginal signs and symptoms score. a. Vulvovaginal Signs: erythema, edema, or excoriation b. Vulvovaginal Symptoms: itching, burning, or irritation c. Scoring Scale: Each score should be objectively defined. The following is an example of an acceptable scale. 0 = none (absent) 1 = mild (slight) 2 = moderate (definitely present) 3 = severe (marked, intense) 13. To establish bioequivalence, the 90% confidence interval of the difference in therapeutic cure rates between the test product and RLD treatment groups at the test-of-cure visit (study Day 21 30) must be within [-0.20, +0.20] for dichotomous variables (cure versus failure), using the PP study population.

Recommended Mar 2011
14. As a parameter for establishing study sensitivity, the test product and RLD should both be statistically superior to placebo (p<0.05) with regard to the therapeutic cure rate at the test-of-cure visit (study Day 21-30), using the mITT study population and Last Observation Carried Forward (LOCF).
15. The sponsor should clearly define the PP, mITT and safety patient populations in the protocol.
16. The accepted PP population used for bioequivalence evaluation includes all randomized patients who had a positive baseline vaginal fungal culture for Candida species, used 1 dose of the assigned study product, and completed the evaluation at the test-of-cure visit within the designated visit window (study Day 21-30) with no protocol violations that would affect the treatment evaluation. The protocol should specify how compliance will be verified, e.g., by the use of patient diaries.
17. The usual mITT population includes all randomized patients who had a positive baseline vaginal fungal culture for Candida species, used at least one dose of study product, and returned for at least one post-baseline visit.
18. The usual safety population includes all randomized patients who received study product.
19. Patients who are discontinued early from the study due to lack of treatment effect at least 3 days after completing treatment should be included in the PP population as treatment failures (i.e., non-responders). Patients discontinued early for other reasons should be excluded from the PP population, but included in the mITT population, using LOCF.
20. Patients who receive or self-administer topical drug therapy for the treatment of vulvovaginal irritation/pruritus after the treatment phase of the study should be analyzed in the mITT and PP populations as a treatment failure.
21. Patients with a negative vaginal fungal culture at baseline should be discontinued from the study and excluded from the PP and mITT populations, but included in the safety population for the safety analyses.
22. All adverse events (AEs) should be reported, whether or not they are considered to be related to the treatment. The report of AEs should include date of onset, description of the AE, and date of resolution. This information is needed to determine if the incidence and severity of adverse reactions is different between the test product and RLD.
23. If the inactive ingredients are different than those contained in the RLD or in significantly different amounts, then the sponsor is to clearly describe the differences and provide information to show that the differences will not affect the safety, efficacy and/or systemic or local availability of the drug.
24. The following Statistical Analysis Method is recommended for equivalence testing for a dichotomous variable (success/failure):

Recommended Mar 2011
Equivalence Analysis Based on the usual method used in OGD for binary outcomes, the 90% confidence interval for the difference in success proportions between test and reference treatment should be contained within [-0.20, +0.20] in order to establish equivalence.