Schwartz, Jill L. MD*; Kovalevsky, George MD†; Lai, Jaim-Jou MPH‡; Ballagh, Susan A. MD†; McCormick, Timothy PhD*; Douville, Karen∥; Mauck, Christine K. MD*; Callahan, Marianne M. MEd*
VAGINAL MICROBICIDE PRODUCTS are currently in development to reduce the transmission of the human immunodeficiency virus (HIV). A class of nonnucleoside reverse transcriptase inhibitors (NNRTIs), the carboxanilides, have been identified, which may have significant therapeutic potential.1 Although originally developed as an oral antiretroviral agent, poor bioavailability prevented UC781’s further development as an oral treatment. As a microbicide, vaginal delivery could allow for sufficient drug delivery to prevent mucosal infection, while limiting systemic drug absorption. UC781 functions as a tight-binder to human immunodeficiency virus-1 (HIV-1) reverse transcriptase (RT)2 and reduces the infectivity of cell-associated virus as well as cell-free virus particles.3
UC781 has been evaluated against a wide range of HIV-1 isolates including laboratory adapted strains, T cell and macrophage tropic isolates, and primary isolates of all major clades (A through G). Consistent with other NNRTIs, UC781 exhibits no activity against HIV-2 isolates, and very little activity against HIV-1 type O isolates, which represent extremely small fractions of worldwide HIV infection. UC781 exerts an inhibitory effect against cells that are chronically infected with HIV-1 and has also been shown to prevent viral transmission to uninfected cells following pretreatment.3 This so-called “memory effect” of UC781 is believed to be a result of the extremely hydrophobic nature of this molecule.
The activity of UC781 against HIV-1 infection has also been evaluated in combination with other HIV-1 RT inhibitors. UC781 has been found to have both additive and synergistic effects4 as well as the ability to restore sensitivity to 3′-azido-3′-deoxythymidine (AZT).5 Unlike other NNRTIs, UC781 activity was not lost after the incidence of a single mutation but required the accumulation of multiple specific mutations to lose efficacy.5 UC781 retained activity against NNRTI-resistant HIV-1 at higher drug concentrations.6
The pharmacokinetic data in multiple animal species indicate that UC781 has generally low oral bioavailability. The effects of UC781 on mammalian cells have been studied in vitro and in cervical explant cultures and have demonstrated that UC781 prevents HIV infection in immune cells7 and inhibits direct infection of mucosal tissue.8,9 The anti-HIV activity of UC781 was equivalent to or better than that of other topical microbicide candidates studied in primary immune cells10 and cervical explant culture.11
This article describes the initial phase I safety study of the safety and acceptability of 0.1%, 0.25%, and 1.0% UC781 gels, compared to the universal placebo gel.
Materials and Methods
This was a single-center, randomized, closed label, phase I study performed in healthy, sexually abstinent women at the CONRAD Clinical Research Center (CRC) at the Eastern Virginia Medical School (EVMS) in Norfolk, VA. Forty-eight women were randomly assigned to 1 of 4 treatment groups: 0.1%, 0.25% and 1.0% UC781 gels, or the universal placebo gel. The primary objective was to examine the effect of UC781 on genital irritation. The study also assessed vaginal health, systemic safety, and acceptability. After a single exposure, each participant was evaluated at the study site, and if cleared for continuation was given the remaining applicators for 5 additional daily exposures. The volunteers, investigators, study sponsor, and analysts were masked as to the product being used by each volunteer to the extent possible. The protocol and informed consents for this study were reviewed and approved by the EVMS Institutional Review Board (IRB) and the study was conducted with the understanding and the signed consent of each participant.
The formulated drug product is composed of UC781 suspended in an aqueous polymer system (carbomer) including preservatives. The universal placebo gel is composed of NaCl, hydroxyethylcellulose, sorbic acid, and NaOH.12,13 All study gels were packaged in prefilled applicators that delivered a volume of 3.5 mL of gel. The absolute amounts of UC781 in 3.5 mL doses of the 0.1%, 0.25%, and 1.0% strength gels were 3.5, 8.75, and 35 mg, respectively. Clinical batches produced for this study were manufactured, processed, packaged, and held under good manufacturing practices (GMP), managed by Biosyn (Huntington Valley, PA).
Selection of Subjects
Volunteers for this study were recruited through existing site databases and local radio and television advertisements. Healthy female volunteers were eligible if they were 18-to 50-years old with regular menstrual cycles, had a normal Pap smear, and were not at risk for pregnancy due to sterilization, hormonal contraceptive use, or abstinence. They agreed to abstain from any vaginal activity including intercourse during the study.
Volunteers were ineligible if they had undergone a hysterectomy; were recently pregnant or breast feeding; were allergic to any component of the study products, N-9, or similar vaginal products; had vaginal candidiasis, bacterial vaginosis (BV), urinary tract infection (UTI) and/or chronic or acute vulvovaginal symptoms; or had been recently diagnosed with a sexually transmitted infection. Volunteers were also ineligible if they had baseline laboratory values ±20% outside of the normal range, had a history of drug or alcohol abuse, recently participated in any other investigational trial or had noniatrogenic deeply disrupted genital epithelium.
Each volunteer was seen in 4 scheduled visits over approximately 1 month (Fig. 1). At the screening visit, signed informed consent, a medical history, dipstick urinalysis, and urine pregnancy test were obtained. A pelvic examination was performed, at which time specimens were collected for Pap smear, if required, wet mount, and N. gonorrhea and C. trachomatis tests. Blood was drawn for complete blood count, chemistry panel, pancreatic enzymes, and UC781 assay.
The second visit was for enrollment and was scheduled to fall within the follicular phase of the menstrual cycle. After a negative urine pregnancy test, colposcopy was carried out.14 A urine dipstick and vaginal wet mount were performed.
Eligible volunteers were randomized to 1 of the 4 treatment groups and were given a prefilled applicator to use at bedtime (Fig. 1). The third and fourth visits were scheduled to take place the day after the first use of the product and the day after the sixth and final use of the product, respectively. At each of the 2 follow-up visits, diary cards and interval medical histories were reviewed by the study staff, and urine pregnancy tests, pelvic examinations, and colposcopies were performed. In addition, at the final visit, the participants completed an acceptability questionnaire, specimens for vaginal wet mount and urine dipstick were collected and blood was drawn for complete blood count, chemistry panel analysis, pancreatic enzymes, and UC781 assay.
Human plasma samples were analyzed for UC781 concentration using a validated HPLC-MS/MS method (Gene Logic, Gaithersburg, MD), with a range of 2.5 to 200.0 ng/mL. The lower level of quantification (LLOQ) for this assay is 2.5 ng/mL. Using weighted linear regression (1/y or 1/y2), the coefficient of determination (r2) for this concentration range is ≥0.99. The percent error of the method is within ±15% (±20% at LLOQ) and the coefficient of variation (CV) is ≤15% (≤20% at LLOQ). All other laboratory studies were conducted in commercial clinical laboratories.
Randomization and Allocation Concealment
The random permuted blocks method was used to generate random allocation sequences to assign 12 participants to each of the 4 treatment groups, using a verified program based on the random function RANUNI in the SAS(r) System (SAS Institute, Cary, NC).
Statistical Analysis of Study Outcomes
The primary endpoint was prespecified as the proportion of women in the product group with signs or symptoms of genital irritation, evaluated separately for (a) events with onsets after first product administration but before the second product administration and (b) events occurring at any time during the 6 days of product use. An estimate of the event proportions and their exact binomial 95% confidence intervals were computed by product group. Adverse experiences during follow-up were tabulated by product group.
The proportions of women experiencing the primary endpoint at any time throughout product use were statistically compared across product groups using the exact Cochran-Armitage trend test (one-sided P value testing for increase in proportion with increasing UC781 concentration). Because of the study’s small sample size, this test has the power to detect only extreme linear trends and should be considered only supplementary to clinical interpretation of these results.
All descriptive and inferential statistical analyses were performed by FHI using the SAS System Version 8 (Cary, NC) and StatXact Version 6 (Cytel, Cambridge, MA).
Enrollment and Demographics
Forty-eight volunteers were enrolled between August 2003 and March 2004. Forty-six (96%) completed the study. Both early discontinuers were in the 0.25% group and discontinued after one product use, one for a medical reason (applicator injury) and the other due to a family problem. There were no women lost to follow-up.
Participant characteristics are summarized in Table 1. There were no apparent differences observed across groups that were likely to affect interpretation of study outcome.
There were 20 complaints of urogenital irritation, almost half of which were from the universal placebo group (Table 2). The most common urogenital events were pelvic pain (cramping, pain, and pressure), genital pruritus, genital pain, and vaginal discharge in that order. Sixteen of the 20 complaints were felt to be mild. The remaining 4 were moderate and included a yeast infection (placebo group, possibly related); a perianal ulcer (placebo group, unrelated); and 2 episodes of pelvic cramping in 1 woman (0.25% group, unrelated). One complaint of breakthrough bleeding occurred in a participant in the placebo group who was on her first month of oral contraceptive pills.
Three participants had a total of 3 noniatrogenic colposcopic findings at baseline (petechiae [placebo group], peeling in the 0.1% and 1% groups), all of which were on the cervical face. None of the findings present at baseline were also present at follow-up and none of the participants who had findings at baseline had additional findings after product use.
Twelve colposcopic findings were seen during follow-up (Table 3). Five findings were first seen after the first gel application (italicized numbers in Table 3). The one laceration was considered a deep epithelial disruption most likely related to the applicator and study gel was discontinued, as per protocol. Only one finding (peeling) was observed again at the final visit.
The 7 remaining findings were first seen at the final visit. Two findings of edema on the vaginal fornix (0.1% and 1% groups) were followed poststudy until they resolved in 8 and 6 days, respectively. One finding of erythema (0.1% group) was resolved at a follow-up visit in 7 days. Four findings of peeling (2 each in the 0.25% and 1.0% groups) were not followed poststudy since peeling is commonly found at baseline.
A post hoc Fisher exact test for colposcopic findings showed no significant difference between placebo versus all UC781 groups combined (two-sided P value = 0.2469) or versus each group separately (two sided P value = 0.5901, 0.1550, and 0.5901), respectively for 0.1%, 0.25%, and 1.0%.
The Z-value for the trend test is −0.8086 (exact one sided P value = 0.7953). The negative Z-value reflects a generally decreasing proportion of events with increasing UC781 concentration (Table 4).
Nonurogenital Adverse Events
There were 17 nonurogenital AEs reported in 13 participants. The most commonly reported event was headache, reported 6 times, with 2 events each in the placebo, 0.25% and 1.0% groups. Eight of the events were blood laboratory values that were outside of the acceptable range but were not considered clinically significant (Table 5). The remaining 3 events were nausea, loss of appetite, and cold.
Two participants had positive wet mounts for yeast at the final visit (one participant in the placebo group was diagnosed with yeast vaginitis and another in the 0.25% group had minimal yeast on wet mount, was asymptomatic and required no treatment). No one developed Trichomonas vaginalis, BV, or a UTI during product use.
Evaluation of UC781 Levels
As expected, there was no detectable level of UC781 in any sample collected at the baseline visit. There were no detectable levels of UC781 in samples collected after treatment with placebo gel, 0.1% UC781 and 0.25% UC781 gel. Two participants in the 1% group had plasma levels that were detectable but <LLOQ (2.5 ng/mL) after treatment. One participant with detectable plasma levels had a colposcopic finding of peeling with no associated adverse events.
All of the women in the study found their product either “easy” or “very easy” to use. Twenty-two (46%) could feel the product; only 5 found the feeling unpleasant (2 in the placebo group and 3 in the 0.25% group). Forty-one (85%) women could not smell the product, but the 7 who could found it unpleasant (2 in the placebo group, 1 in the 0.1% group, 3 in the 0.25% group, and 1 in the 1.0% group).
Forty-four (92%) reported very slight to moderate product leakage. Of these, 8 (3 in the placebo group, 2 in the 0.1% group, 2 in the 0.25% group, and 1 in the 1.0% group) reported that leakage would prevent them from using the gel in the future.
All of the participants, except for one in the 0.1% UC781 group who indicated that she would prefer an oral agent, would buy the product if they needed protection against HIV and if the product was proven effective. Most of the women who had used spermicide in the past indicated that the study product they used was the same as or better.
This is the first phase I clinical safety study of UC781. UC781 at three concentrations applied once daily for 6 days was well tolerated in this group of sexually abstinent women. The occurrence of urogenital irritation with UC781 was not noticeably different than in another phase I safety study of an antiretroviral microbicide candidate.15
Some evidence of urogenital irritation was common in all treatment groups and was most often mild and transient. The nonsignificant negative Z-value observed in this study implies a generally decreasing proportion of urogenital events with increasing UC781 concentration. However, the inverse relationship of dose to events is most likely due to the more frequent occurrence of urogenital events in the placebo group and should not be interpreted as evidence that UC781 has a protective effect.
Colposcopic findings were infrequent in the placebo group (8%) and those in the 3 treatment groups (24% to 42%) are consistent with observations in other small microbicide safety studies.15,16 However, direct comparisons between products are difficult because of small sample sizes typical of preliminary safety studies and variations in reporting between studies. No specific observed colposcopic finding has been conclusively linked to an increased risk of HIV acquisition. Peeling and erythema are frequently found at baseline before product application and are not felt to pose a significant safety risk. Edema, which may indicate underlying inflammation, has been observed with the use of nonoxynol-9 products,17 and was observed in the vaginal fornix of 2 study subjects using UC781. Deep disruption of the epithelium is felt to carry the greatest risk because grossly observed epithelial breaches were associated with HIV seroconversion in a study of nonoxyol-9 (N-9).18 Another microbicide candidate, Carraguard, has reached a phase III trial without the use of routine colposcopy in early studies; the investigators questioned the role of routine colposcopy given the difficulty interpreting the significance of findings in preliminary safety studies and the expense of the procedure.19
To understand the full significance of colposcopic findings in the evaluation of vaginal products, the correlation between colposcopic findings such as edema and other potential biomarkers of inflammation including soluble and cellular markers and biopsy, needs to be further explored. Such work is underway at CONRAD. To characterize inflammatory profiles of 2 unsuccessful microbicide candidates (N-9 and cellulose sulfate) and a placebo gel, standard safety surveillance and collection of inflammatory biomarkers in cervicovaginal tissues and lavages will be conducted following vaginal application.
Given that topical application is expected to limit systemic absorption and minimize side effects and the development of resistance, the finding of detectable plasma levels in 2 of 12 participants using 1% UC781 gel indicates that 1% is not the optimal concentration, especially if dosing regimens are more frequent than once daily.
Five safety/PK studies of UC781 are currently underway in women and men, all utilizing UC781 concentrations less than 1%, with twice-daily dosing in some studies and all involving careful monitoring of exposed epithelium.
1. Balzarini J, Brouwer WG, Dao DC, et al. Identification of novel thiocarboxanilide derivatives that suppress a variety of drug-resistant mutant human immunodeficiency virus type 1 strains at potency similar to that for wild-type virus. Antimicrob Agents Chemother 1996; 40:1454–1466.
2. Barnard J, Borkow G, Parniak MA. The thiocarboxanilide nonnucleoside UC781 is a tight-binding inhibitor of HIV-1 reverse transcriptase. Biochemistry 1997; 36:7786–7792.
3. Borkow G, Barnard J, Nguyen TM, et al. Chemical barriers to human immunodeficiency virus type 1 (HIV-1) infection: Retrovirucidal activity of UC781, a thiocarboxanilide nonnucleoside inhibitor of HIV-1 reverse transcriptase. J Virol 1997; 71:3023–3030.
4. Buckheit RW, Snow MJ, Fliakas-Boltz V, et al. Highly potent oxathiin carboxanilide derivatives with efficacy against nonnucleoside reverse transcriptase inhibitor-resistant human immunodeficiency virus isolates. Antimicrob Agents Chemother 1997; 41:831–837.
5. Borkow G, Arion D, Wainberg MA, et al. The thiocarboxanilide nonnucleoside inhibitor UC781 restores antiviral activity of 3′-azido-3′-deoxythymidine (AZT) against AZT-resistant human immunodeficiency virus type 1. Antimicrob Agents Chemother 1999; 43:259–263.
6. Hossain MM, Parniak MA. In vitro microbicidal activity of the nonnucleoside reverse transcriptase inhibitor (NNRTI) UC781 against NNRTI-resistant human immunodeficiency virus type 1. J Virol 2006; 80:4440–4446.
7. Van Herrewege Y, Michiels J, Van Roey J, et al. In vitro evaluation of nonnucleoside reverse transcriptase inhibitors UC–781 and TMC120-R147681 as human immunodeficiency virus microbicides. Antimicrob Agents Chemother 2004; 48:337–339.
8. Zussman A, Lara L, Lara H, et al. Blocking of cell-free and cell-associated HIV-1 transmission through human cervix organ culture with UC781. AIDS 2003; 17:653–661.
9. Fletcher P, Kiselyeva Y, Wallace G, et al. The nonnucleoside reverse transcriptase inhibitor UC781 inhibits human immunodeficiency virus type 1 infection of human cervical tissue and dissemination by migratory cells. J Virol 2005; 79:11179–11186.
10. Dezzutti CS, James VN, Ramos A, et al. In vitro comparison of topical microbicides for prevention of human deficiency virus type 1 transmission. Antimicrob Agents Chemother 2004; 48:3834–3844.
11. Cummins JE, Guarner J, Flowers L, et al. Preclinical testing of candidate topical microbicides for anti-human immunodeficiency virus type 1 activity and tissue toxicity in a human cervical explant culture. Antimicrob Agents Chemother 2007; 51:1770–1779.
12. Tien D, Schnaare R, Kang F, et al. In vitro and in vivo characterization of a potential universal placebo designed for use in vaginal microbicide clinical trials. AIDS Res Hum Retroviruses 2005; 21:845–853.
13. Schwartz JL, Ballagh SA, Kwok C, et al. Fourteen day safety and acceptability study of the universal placebo gel: A randomized double blind phase I safety study. Contraception 2007; 75:136–141.
14. World Health Organization and CONRAD Manual for the standardization of colposcopy for the evaluation of vaginally administered products, Update 2000: Geneva, Switzerland/Arlington, VA: WHO/CONRAD, 2000.
15. Mayer KH, Maslankowski LA, Fai F, et al. Safety and tolerability of tenofovir vaginal gel in abstinent and sexually active HIV-infected and uninfected women. AIDS 2006; 20:543–551.
16. Van Damme L, Wright A, Depraetere K, et al. A phase I study of a novel potential intravaginal microbicide, PRO 2000, in healthy sexually inactive women. Sex Transm Infect 2000; 76:126–130.
17. Watts H, Rabe L, Krohn MA, et al. The effects of three nonoxynol-9 preparations on vaginal flora and epithelium. J Infect Dis 1999; 180:426–437.
18. Van Damme L, Ramjee G, Alary M, et al.; COL–1492 Study Group. Effectiveness of COL-1492, a nonoxynol-9 vaginal gel, on HIV-1 transmission in female sex workers: A randomised controlled trial. Lancet 2002; 360:971–977.
19. Coggins C, Blanchard K, Alvarez F, et al. Preliminary safety and acceptability of a carrageenan gel for possible use as a vaginal microbicide. Sex Transm Infect 2000; 76:480–483.