Phase 3 Safety Comparisons for 1.0% Azithromycin in Polymeric Mucoadhesive Eye Drops versus 0.3% Tobramycin Eye Drops for Bacterial Conjunctivitis

Abstract

purpose. To compare the safety and tolerability of 1.0% azithromycin in a polymeric mucoadhesive delivery system with 0.3% tobramycin ophthalmic solution for the treatment of bacterial conjunctivitis.

methods. This study was a prospective, randomized, active-controlled, double-masked, phase 3 trial conducted from August 6, 2004, to October 6, 2005, at 47 sites. Subjects with a clinical diagnosis of bacterial conjunctivitis were randomly assigned to receive either 1% azithromycin in DuraSite (AzaSite; InSite Vision, Alameda, CA) (n = 365) or 0.3% tobramycin (n = 378). Both groups received masked medication four times daily for 5 days, but participants received an active dose of 1% azithromycin in DuraSite only twice a day on days 1 and 2 and daily on days 3 to 5. Conjunctival cultures were taken, and ocular signs and symptoms were evaluated at baseline and at two follow-up visits.

results. A total of 743 patients were randomized, and 710 (96%) completed the trial. Both study medications were well tolerated. The most frequently observed ocular adverse events in the azithromycin group were eye irritation (1.9%), conjunctival hyperemia (1.1%), and worsening bacterial conjunctivitis (1.1%). These rates compared favorably with those obtained with tobramycin. Rates of microbial eradication (an efficacy parameter) and bacterial infection recurrence (a safety parameter) were the same in both groups.

conclusions. This is the first report of the safety and tolerability of a commercially manufactured preparation of azithromycin for ophthalmic use. Azithromycin 1% in DuraSite is safe and can be administered in a regimen of less frequent doses than can tobramycin, while producing an equivalent clinical outcome. The formulation is well tolerated in patients over the age of 1 year for the eradication of bacteria commonly associated with conjunctivitis. (ClinicalTrials.gov number, NCT00105469.)

Bacterial conjunctivitis is a common infectious condition. Infections are usually self-limiting, but anti-infective treatment is used to shorten the course of disease, limit contagion, and reduce the risk of complications. Gram-positive staphylococcus species are the most common causative pathogens followed by Streptococcus pneumoniae and Haemophilus influenzae. ^{1 2} However, H. influenzae is the most common cause in the pediatric patient. Other atypical species may also be present. Most cases of bacterial conjunctivitis in the adult and pediatric ophthalmic settings can benefit from topical anti-infective treatment. Although several antimicrobial agents are effective against bacterial conjunctivitis, fluoroquinolones are considered a cornerstone of antibacterial therapy. These topical anti-infectives are not recommended in infants and children less than 1 year of age. Systemic use in children is associated with drug class side effects. ³ The American Academy of Pediatrics recommends that fluoroquinolone use be restricted to situations in which there is no safe and effective alternative. ⁴ Several topical fluoroquinolones are used today to treat acute ocular surface infections in children and adults, despite inconvenient dose regimens and increasing bacterial resistance to the drug class. ⁵  

Azithromycin is a macrolide-class anti-infective in the azalide subclass. It is synthesized from erythromycin and possesses a well-known safety and tolerability profile in both oral and intravenous forms. ^{6 7} In ophthalmology, azithromycin is more familiarly used in an oral form to eradicate trachoma. ⁸ The safety profile of a new topical ocular solution of 1.0% azithromycin in DuraSite, a polymeric mucoadhesive delivery system (InSite Vision, Alameda, CA), was evaluated in adult and pediatric participants with bacterial conjunctivitis and compared with that of tobramycin 0.3% ophthalmic solution, USP. DuraSite has been used safely in ophthalmic applications before, but this is the full report of safety data from a phase 3 trial in which the system was used. Parts of these data have been reported in abstract form (Protzko E et al. IOVS 2006;47:ARVO E-Abstract 4958). In addition to the evaluation of ocular adverse events, the safety profile data presented herein also include factors that are important to the successful therapeutic intervention of infective bacterial conjunctivitis, including patients’ adherence to the dose regimen, drug tolerance, and microbial spectrum. 

Methods

This research was conducted according to the tenets of the Declaration of Helsinki. Study investigators were approved by the governing institutional review boards for each site. Informed consent or parental assent was obtained from the participants after explanation of the nature and possible consequences of the study. The phase 3 trial was a prospective, randomized, active-controlled, double-masked study that was conducted at 47 sites from August 6, 2004, to October 6, 2005. The study included 743 participants who ranged in age from 1 to 96 years who had a diagnosis of bacterial conjunctivitis of less than 3 days’ onset. The safety sample was composed of all participants who received a single dose of study drug (n = 743). Statistics were calculated for the intent-to treat population, which included all participants in the safety sample who returned to the clinic for at least one post-dose follow-up visit (n = 735). 

The primary assessment of safety and tolerability was based on measurements of adverse events, visual acuity, indirect ophthalmoscopy, and biomicroscopy. The extent of recurrence of the original pathogen and presence of novel bacteria at different time points in the study were also used as a measure of safety. The study design is shown in Figure 1 . Participants were randomized and instructed to instill masked medication four times daily (at 4–6 hour intervals) on days 1 through 5. To mask this study, subjects were given four bottles for each day. On days 1 and 2 only two of the bottles in the 1% azithromycin in DuraSite arm contained the active azithromycin formulation. On days 3 through 5, only one of four bottles contained 1% azithromycin in DuraSite. For masking, the intermediate doses contained the formulation vehicle. 

Disposition of Study Participants

A total of 747 participants were enrolled. The records for four participants were lost in centers that were damaged during Hurricane Katrina in 2005. Of the 743 remaining participants, 710 (95.6%) completed the trial: 94% (343/365) of those in the 1% azithromycin in the DuraSite group and 97.1% (367/378) of those in the tobramycin group. 

Treatment Discontinuation

The two treatment groups had similar overall discontinuation rates. Thirty-three participants were terminated from the study before completion (Table 1) . Of the early terminations, 17 were as a result of adverse events and the remaining 16 were classified as one of the following: protocol violation, withdrawn consent, lost to follow-up, or lack of efficacy. 

Study Participant Demographics

Baseline demographics were captured and statistics calculated for the intent-to-treat group. In the overall study population, participants ranged in age from 1 to 93 years. In the group that received 1% azithromycin in DuraSite, 36.1% were pediatric (range, 1–11 years) and 59.2% were adult (17–87 years). Most of the participants identified themselves as white. More than 5% stated that they were either African American or Hispanic. More than half of the participants had brown eyes. There were no significant differences between the treatment groups in sex, race, or eye color (Table 2) . 

Results

Adverse Events

Adverse event data were listed and summarized according to treatment group, body system, investigator opinion concerning the relation to drug therapy, and severity. There was no significant difference in the frequency of adverse events between the two treatment groups (Table 1) . Among all adverse events reported, 3% (11/365) were deemed treatment-related in the 1% azithromycin in DuraSite group and 5.6% (21/378) in the tobramycin group. The most frequently observed ocular adverse events in the overall study population were eye irritation (1.9%), conjunctival hyperemia (1.1%), and worsening conjunctivitis (1.1%, Table 3 ). 

Visual Acuity

Eligible study participants were required to have a best corrected visual acuity (BCVA) of ≥20/100 in both eyes at baseline (day 1/study visit 1). Visual acuity was analyzed again in each eye at visit 2 (day 3 + 1) and 3 (day 6 + 1). Evaluations were conducted using the Snellen scale for visual acuity. Lea symbols were used with children who could not be tested with the Snellen chart. At each participant’s final visit, the score of the treated eye with the greater decline in BCVA was captured in statistical analysis. In this study, a clinically significant change was considered to be a change in visual acuity ≥3 lines. The percentage of participants with a clinically significant decline in visual acuity of 3 lines or more at any visit (schedule or unscheduled) was 0.8% in either treatment arm. More than 96% of participants had no change in visual acuity at any visit during the course of treatment. 

Biomicroscopy and Ophthalmoscopy

Slit lamp biomicroscopy was performed at all visits. Ophthalmoscopy of the undilated fundus was performed at baseline (day 1) and visit 3 (day 6 + 1). Few subjects experienced any worsening of ophthalmic signs. The most frequent treatment-emergent outcome was swelling of the eyelid, which was seen in 12 (3.3%) participants in each treatment group (Table 4) . Other findings in the conjunctiva, lids, and cornea were equally distributed at relatively low frequencies in both treatment groups. 

Pathogens at Study Exit

The safety of the 5-day treatment regimen on the development of new bacterial infections and the susceptibility pattern of bacteria populations was evaluated. Conjunctival samples from all intent-to-treat patients were cultured at study entry. 

In the first test, conjunctival cultures were evaluated at study exit to detect the occurrence of new pathogens after the initiation of treatment. Pathogens that were identical with those found at study entry were not captured in this data set. The rate of new pathogens was very small. In the intent-to-treat population new bacteria were observed in 17 (4.6%) of 365 participants who received 1% azithromycin in DuraSite and in 12 (3.2%) of 378 participants who received tobramycin. Because participants in the overall intent-to-treat population may not have taken all doses, may not have returned for some follow-up visits, or may have presented with positive bacterial cultures at study entry, we also examined the rate of new pathogens in the per-protocol group. This population met all entry requirements, took all doses, and returned for each follow-up visit. Among the per-protocol groups the rates of new pathogens in participants who received 1% azithromycin in DuraSite and those who received tobramycin were similar. (Table 5) . 

The second measure examined the nature of organisms with respect to Gram’s stain present at entry and compared that to the bacterial pathogens that were present in levels above pathologic threshold at study exit. This measure was undertaken to detect whether there were shifts in the frequency of commonly observed bacteria during the trial. All treated eyes were examined. There was a wide range of pathologic bacteria species found at the initial visit. Gram-positive streptococci and Gram-negative Haemophilus were each found in approximately one third of the cases. 

In the per-protocol group (1% azithromycin in DuraSite, n= 159; 0.3% tobramycin, n= 157) the causative pathogens that were detected in five or more participants per treatment group were: H. influenzae (42.8% in azithromycin and 36.3% in tobramycin group), S. pneumoniae (39.6% in the azithromycin and 42.7% in the tobramycin group), Staphylococcus aureus (12.6% in the azithromycin and 14.6% in the tobramycin group), and Staphylococcus epidermidis (3.1% in the azithromycin and 3.2% in the tobramycin group). 

At visit 3, the most frequently observed bacteria in the 1% azithromycin treatment group were S. pneumoniae (2.5%) and H. influenzae (2.2%). The nature and ratio of the bacteria in the tobramycin treatment group were similar. Both rates were consistent with the proportions observed at study entry and suggest that the treatments were equally capable of eradicating the predominant Gram-negative and Gram-positive pathogens in this group. 

Discussion

The safety and tolerability of the 5-day regimen of topical 1% azithromycin in DuraSite was compared to that of 0.3% tobramycin ocular solution. The new formulation is a preserved gel-forming drop composed of 1% azithromycin in DuraSite. On instillation 1% azithromycin in DuraSite persists on the ocular surface longer than conventional aqueous eye drops. DuraSite, a polymeric mucoadhesive ocular delivery system, affords this property and enhances the retention of azithromycin in the conjunctiva. When formulated with a solution of azithromycin, DuraSite produces a sustained-release gel that flows like an eye drop while delivering the long-term coverage of an ointment. In this study as well as others, the DuraSite delivery vehicle did not cause blurriness (Keller N et al. IOVS 1993;34:ARVO Abstract 3907). ⁹ More than 96% of participants experienced no changes in visual acuity at any time during the treatment. 

Topical ocular preparations of moxifloxacin and gatifloxacin are well known for their activity against Gram-positive, Gram-negative, and atypical bacteria. However, the prescribed dosage of these drugs ranges between three and eight times per day for 7 days. The frequent instillation of these drops is necessary to obtain critical antibiotic concentrations in the eye. The DuraSite delivery vehicle extends the residence time of azithromycin in the eye and permits a less-frequent dose regimen. 

Approximately one-third of all participants were classified as pediatric, half were between the ages of 17 and 64 years, and 5% were older than 65 years. Overall, 95.6% of participants completed the 5-day course of antibiotics. According to their diaries, participants instilled more than 98% of the prescribed medication. Studies have shown that patients comply more readily with dosage instructions when the regimen is simple and a complete course of therapy requires fewer doses. ¹⁰  

In a literature review of 76 studies, Taylor et al. ¹¹ assert that a medicine that requires a once-daily dose is more likely to be taken, and taken correctly, than one that must be taken in four doses over the same day. Less frequent doses and shorter courses of therapy have the most positive influence on adherence. The data in this clinical trial show that the patients complied with the shorter course of therapy with the 1% azithromycin formulation, and that it was safe in preventing the recurrence of infection. Although real-world compliance data are not yet available with 1% azithromycin in the polymeric delivery system for acute ocular infection, it could be argued that in addition to giving physicians a new treatment option for eradicating bacteria, its simpler dose regimen and shorter course of treatment could improve patients’ compliance, reduce treatment failures, and limit the development of serious sequelae such as ulcers. 

With the clinical experience obtained across a broad range of ages in this trial, there was a remarkably low incidence of ocular adverse events and no difference between treatment groups in the incidence of treatment-related adverse events. Both drops are preserved with benzalkonium chloride. The azithromycin and tobramycin formulations contain 0.003% and 0.01% benzalkonium chloride, respectively. The preservative in some eye drops can contribute to ocular irritation. Package label documentation of the similarly preserved (0.005% benzalkonium chloride) ocular preparation of gatifloxacin (Zymar; Allergan, Irvine, CA) reported that the most frequent ocular adverse events of conjunctival irritation, increased lacrimation, keratitis, and papillary conjunctivitis occurred in approximately 5% to 10% of patients in the overall study population, compared with the 1.9% to 3.3% in this study. ¹² Similarly, ocular adverse events were reported in 9% of patients in the 0.5% levofloxacin clinical trial and included transient burning (2.4%) and decreased vision (2.4%). ¹³ Although the specific adverse events that were captured in these different studies may have been different, the ocular adverse events ratio in the 1% azithromycin trial is one of the lowest among commercially available eye drops. 

The incidence of bacteria that could be cultured from the ocular surface during treatment and at study exit was low in all groups. Additional work is needed to evaluate whether the bacteria that were cultured were resistant to either treatment. The population examined in these safety studies was the intent-to-treat group, which included all patients who had clinical signs of bacterial conjunctivitis and who received a single drop of study medication. Some participants may not have taken all the scheduled doses or presented at study entry with positive ocular cultures. Therefore, the study is not an accurate reflection of resistance. The data indicate that the participants who adhered to the 5-day regimen of 7 drops of 1% azithromycin in DuraSite were just as protected as those who took 20 drops of 0.3% tobramycin. 

At study exit the nature of bacteria at pathologic threshold was consistent with the most prevalent bacteria in community-acquired bacterial conjunctivitis. Gram-positive S. pneumoniae and Gram-negative H. influenzae represented about a third of bacterial types cultured at study entry. Treatment with either 1% azithromycin or tobramycin resulted in a greater than 15-fold reduction in these bacteria. 

This safety analysis confirms that 1% azithromycin in DuraSite administered twice daily for 2 days and then once a day for 3 days achieved a safety profile similar to tobramycin at four doses a day for 5 days. The 5-day regimen with 1% azithromycin was well tolerated in patients of age 1 year and older. The safety profile compares favorably with 0.3% tobramycin and other commercially available eye drops. No significant differences in the overall incidence of adverse events and other safety variables were detected between the treatment groups. 

Appendix 1

The AzaSite Bacterial Conjunctivitis Active-controlled Clinical Study Group

John Alder, Pranav Amin, James Aquavella, Lourdes Arellanes-Garcia, Maria Avila-Aguero, Alan Bauman, Gregg Berdy, Brian Bowe, William Bray, Delmar Caldwell, Maria Mercedes Alba Castrejon, Alexander Chang, Jason Chiappetta, Kathryn Colby, John Cowden, Jung Dao, Gerard D’Aversa, William Davitt, Peter S. Dawson, Chad Denison, Peter Donshik, Sima Doshi-Carnavale, Steven Elliott, Jon Fishburn, Kerry Hagen, Ben Hasty, Richard Hector, Evan Held, Harold Helms, Michael S. Insler, Elisa Jimenez-Fonseca, Anthony P. Johnson, Randy Katz, Barry Katzman, Fred Kurata, Michael Lappin, Roberto Leal-Leyva, Norm Levy, Steven Lichtenstein, Andrew Lipka, Kathia Luciani-Chiu, Jonathan I. Macy, Laura Muller, Terrence O’Brien, Kevin O’Neal, Matthew Paul, Henry Perry, M. Millicent Peterseim, Christodoulos M. Petras, Michael Phillips, Luis Carlos Porras-Mendoza, Eugene Protzko, John Puglisi, Peter Rapoza, Robert Reidy, Richard Shugarman, Alan Shuster, Claudio Slon-Hitti, Steve Spector, Emil Stein, Marixcel Suarez Fernandez, Michael Tepedino, Robert A. Thomas, Mariano Tovar Rivera, Roberto Javier Vasquez-Hernandez, Irving Weinberger, Julius W. Welborn, Robert Wiggins, Craig Wilkerson, and Richard Yee. 

⁶  Group members are listed in the Appendix.

 Presented in part at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 2006.

 Submitted for publication November 27, 2006; revised January 14, 2007; accepted May 23, 2007.

 Disclosure: E. Protzko, None; L. Bowman, Insite Vision (E); M. Abelson, Insite Vision (C); A. Shapiro, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Study design and dosage scheme.

Table 1.

 

Disposition of the Enrolled Subjects

	1% Azithromycin in DuraSite (n = 365)	0.3% Tobramycin (n = 378)	Total (n = 743)
Randomized	365 (100.0%)	378 (100.0%)	743 (100.0%)
Completed	343 (94.0%)	367 (97.1%)	710 (95.6%)
Discontinued	22 (6.0%)	11 (2.9%)	33 (4.4%)
Adverse event	9 (2.5%)	8 (2.1%)	17 (2.3%)
Protocol violation	4 (1.1%)	0	4 (0.5%)
Withdrew consent	2 (0.5%)	3 (0.8%)	5 (0.7%)
Lost to follow-up	1 (0.3%)	0	1 (0.1%)
Lack of efficacy	2 (0.5%)	0	2 (0.3%)
Other	4 (1.1%)	0	4 (0.5%)
Treatment population
Safety	365 (100.0%)	378 (100.0%)	743 (100.0%)
Intent-to-treat	360 (98.6%)	375 (99.2%)	735 (98.9%)
Per protocol*	159 (43.6%)	157 (41.5%)	316 (42.5%)
Efficacy evaluable, †	137 (37.5%)	131 (34.7%)	268 (36.1%)

*  Per protocol population was all randomized participants who had no major protocol violations.

†  Efficacy evaluable data set included all per-protocol subjects whose protocol violations did not affect the efficacy data.

Table 2.

 

Study Demographics: Intent-to-Treat Group

	1% Azithromycin in DuraSite (n = 360)	0.3% Tobramycin (n = 375)	P
Age, y
Mean age ± SD	26.2 ± 21.48	27.9 ± 21.73	0.289
Median age	22.5	26
Range	1–87	1–93
0–11	130 (36.1%)	125 (33.3%)
12–16	17 (4.7%)	20 (5.3%)
17–64	194 (53.9%)	204 (54.4%)
≥65	19 (5.3%)	26 (6.9%)
Sex
Male	158 (43.9%)	160 (42.7%)	0.766
Female	202 (56.1%)	215 (57.3%)
Race
Caucasian	239 (66.4%)	242 (64.5%)	0.306
African American	25 (6.9%)	40 (10.7%)
Asian or Pacific Islander	19 (5.3%)	12 (3.2%)
Hispanic	69 (19.2%)	75 (20.0%)
Native American or Alaskan	3 (0.8%)	1 (0.3%)
Other	5 (1.4%)	5 (1.3%)
Eye color
Brown	183 (50.8%)	192 (51.2%)	0.493
Blue	97 (26.9%)	108 (28.8%)
Green	30 (8.3%)	19 (5.1%)
Hazel	39 (10.8%)	45 (12.0%)
Other	11 (3.1%)	11 (2.9%)
Iris color (hue)
Dark	192 (53.3%)	203 (54.1%)	0.801
Hazel	39 (10.8%)	45 (12.0%)
Light	129 (35.8%)	127 (33.9%)

Table 3.

 

Summary of the Most Frequent Ocular Adverse Events

	1% Azithromycin in DuraSite (n = 365)	0.3% Tobramycin (n = 378)	Total (n = 743)
Eye irritation	7 (1.9%)	4 (1.1%)	11 (1.5%)
Conjunctival hyperemia	4 (1.1%)	4 (1.1%)	8 (1.1%)
Worsening bacterial conjunctivitis	4 (1.1%)	8 (2.1%)	12 (1.6%)

Table 4.

 

Summary of Treatment-Related Outcomes That Worsened in Treated Eyes

Result	1% Azithromycin in DuraSite (n = 365)	0.3% Tobramycin (n = 378)
Lids swelling	12 (3.3%)	12 (3.2%)
Conjunctiva chemosis	8 (2.2%)	10 (2.6%)
Conjunctiva, other	8 (2.2%)	15 (4.0%)
Lids, other	8 (2.2%)	4 (1.1%)
Cornea staining erosion	7 (1.9%)	6 (1.6%)

Table 5.

 

Summary of Patients with New Bacteria Present above the Pathologic Threshold at Study Exit

	Per Protocol	Intent-to-Treat
1% Azithromycin in DuraSite	12 (n = 159)	17 (n = 365)
0.3% Tobramycin	10 (n = 157)	12 (n = 378)

The authors thank James G. Shook and Ping Hsu for the development of the clinical trial data and Cheryl Rowe-Rendleman for technical contributions to the manuscript.