Intranasal Surfactant for Acute Otitis Media: A Randomized Trial

We conducted this trial from February 2019 to February 2020 at the University of Pittsburgh Medical Center Children’s Hospital of Pittsburgh Primary Care Center and Express Care and at affiliated pediatric practices in Pittsburgh, Pennsylvania, representing urban and suburban demographics. The protocol was approved by the institutional review board, and written informed consent was obtained from a parent of each enrolled child. The authors attest that the study was performed in accordance with the protocol, including its statistical analysis plan. The study was sponsored by Novus Therapeutics, Inc, and registered in ClinicalTrials.gov (NCT03818815). Novus Therapeutics provided study product and statistical support for data analysis.

Eligible children 6 to 24 months of age were required to have AOM on the basis of onset of symptoms within the preceding 48 hours (ie, a score of 5 or more on the 5-item semiquantitative Acute Otitis Media–Severity of Symptoms [AOM-SOS] scale version 5.0)^9,10  and an intact tympanic membrane with mild bulging with otalgia or marked tympanic membrane erythema, or moderate to severe bulging. The AOM-SOS scale consists of 5 discrete items: tugging of ears, crying, difficulty sleeping, irritability, and fever. Parents are asked to rate these symptoms, as compared with the child’s usual state, as “none,” ‘almost none,” “a little,” “some,” “a lot,” or “an extreme amount” with corresponding scores of 0, 1, 2, 3, 4, and 5. Thus, total scores range from 0 to 25, with higher scores indicating greater severity of symptoms. Parents of eligible children also were required to have a smartphone or Internet access to receive electronic surveys. We excluded children who had a tympanic membrane perforation or tympanostomy tubes; were allergic to amoxicillin or cephalosporins; had received an antimicrobial agent within the previous 72 hours; had used intranasal treatments other than saline within the previous 7 days; had a history of immunodeficiency disorders, craniofacial abnormalities that may interfere with ET function, disorders with decreased mucociliary clearance or higher viscosity of the mucous, or clinically relevant blockage of 1 or both nasal passages; or those whose parents were employed by the sponsor or investigator.

This was a phase 2a, single center, double-blind, randomized, placebo-controlled, parallel group study to assess the safety, tolerability, and efficacy of 20 mg per day intranasal OP0201 as adjunct therapy to oral antimicrobial treatment of children aged 6 to 24 months (inclusive) with AOM (Fig 1). Children who met eligibility criteria were randomly assigned in a 1:1 ratio within strata (age 6 to 11 months or 12 to 24 months, and whether they were exposed to ≥3 children for ≥10 hours per week) to receive a total of 20 intranasal doses of OP0201 (10 mg administered twice daily for 10 days) or placebo, as adjunct treatment to oral antimicrobial agents (amoxicillin-clavulanate 90/6.4 mg/kg in 2 divided doses for 10 days). The study treatment included a pressurized canister with a securely attached metering valve, where OP0201 or placebo was contained, an actuator device into which the canister is seated, and an angled tip on the device that delivered the product into the nostrils. Placebo contained the propellant without any active ingredients (HFA 134a). We trained parents to administer study treatment and gave the first dose under supervision by study personnel.

Participants were managed for 30 days, including 4 clinic visits: visit 1 (day 1), screening and enrollment and initiation of the study treatment and oral antimicrobial agents; visit 2 (day 4 [+2]); visit 3 (day 12 [+2]); and visit 4 (day 28 [±2]). Parents and caregivers of children enrolled completed an electronic daily diary on all days when study treatment was administered (∼10 days) to record twice-daily administration of the oral antimicrobial and study treatment and to complete the AOM-SOS scale. At visit 3, parents and caregivers completed a questionnaire on their experience with using the delivery device. We scheduled interim evaluations if a parent or caregiver notified investigators that their child experienced no improvement, significantly worsened, or had recurrence of signs and symptoms of AOM, or developed symptoms that may have been related to study treatment.

All 7 study clinicians successfully completed an otoscopic validation program before the start of the study.^11,12  At each study visit, assessments included a physical examination (including visual examination of the nasopharynx and oropharynx) and measurement of vital signs, pneumatic otoscopy, and endoscopic examination with capture of an image of the tympanic membrane and tympanometry when possible. Adverse events and concomitant medication use were recorded. At visit 2 (day 4 [+2]), we specifically assessed for bulging of the tympanic membrane. At the end-of-treatment visit 3 (day 12 [+2]), we assessed (1) treatment failure as evidenced by the presence of moderate to severe bulging of the tympanic membrane or mild bulging of the tympanic membrane with recent (<48 hours) onset of ear pain (otalgia) or intense erythema or the tympanic membrane, (2) rate of persistence of middle-ear effusion, and (3) AOM-SOS scores during the 10 days of treatment. Children categorized as exhibiting treatment failure were prescribed an additional treatment course of amoxicillin-clavulanate for 10 days or ceftriaxone administered intramuscularly at a dose of 75 mg/kg and repeated 48 hours later.

All endpoints were prespecified. There were 2 primary efficacy endpoints: (1) no bulging of the tympanic membrane (with or without symptoms) at visit 2 on day 4 [+2], and (2) no middle-ear effusion at visit 3 on day 12 [+2]. Secondary efficacy endpoints included (1) absence of bulging of the tympanic membrane at visits on days 12 [+2] and 28 [±2]; (2) absence of middle-ear effusion at visits on days 4 [+2] and 28 [±2]; (3) type A (normal) tympanogram at visits on days 4 [+2], 12 [+2], and 28 [±2]; (4) complete or near complete resolution of symptoms (AOM cure), defined as an AOM-SOS score ≤2, assessed at visits 2, 3, and 4; (5) ≥50% reduction from baseline in the AOM-SOS score assessed at visits on days 4 [+2], 12 [+2], and 28 [±2]; (6) change from baseline in AOM-SOS score over the 10-day treatment period by parent or caregiver diary data; (7) clinical treatment failure at visit 3 (day 12 [+2]); and (8) recurrence of AOM. We also assessed the parent’s experience with using the device.

We assessed safety by incidence, seriousness, severity, and relationship to study treatment (OP0201 or placebo) of emergent adverse events; and by shifts from baseline in vital signs (normal, abnormal), physical examination (normal, abnormal), and examination of the nasopharynx and oropharynx (normal, abnormal) for local effects.

We estimated that randomizing 140 children (1:1) in this phase 2a study would provide 93% power to detect a treatment effect of at least 0.25 difference in proportions in at least 1 of the 2 primary endpoints (no bulging of the tympanic membrane at day 4 [+2], and no middle-ear effusion at day 12 [+2]), on the basis of a 2-sided Pearson χ² test of a 2-sided hypothesis for each endpoint and controlling for type 1 error rate of 0.05 using the step-up Hochberg procedure.¹³  We assumed that 40% of placebo-treated participants would not have a bulging tympanic membrane at day 4 and that 30% of placebo-treated participants would not have middle-ear effusion at day 12 [+2]. We also assumed a 10% drop out rate with multiple imputation of missing values.

The primary efficacy analysis was prespecified to be done using the modified intent-to-treat (mITT) population, defined as participants who were assigned randomly; excluding the first 3 participants dosed as there was a known device malfunction for these participants; and with summaries and analyses based on randomized treatment assignment regardless of treatment received. We based the post hoc efficacy analyses presented in this report on the subset of children who completed the assessments in the mITT population, whereby for each endpoint, we included only participants with complete data, and used slightly more inclusive analysis windows than those that were prespecified. We summarized safety parameters using the safety population, defined as children who received at least one intranasal spray of study treatment. For safety, we based summaries on treatment received, regardless of randomized treatment assignments.

We summarized categorical variables (including binary outcomes) as frequencies and percentages and analyzed them using logistic regression to control for covariates, including the randomization stratification variables. Continuous variables were summarized by numbers of observations, means, measures of variances, percentiles, and ranges and were analyzed by using general linear models to control for covariates.

We randomly assigned a total of 103 children of the 140 originally planned because of sponsor’s budgetary constraints. For the initial 3 children, a device malfunction precluded delivery of an appropriate dose of study medication; the mITT population excluded these 3 children and therefore included 100 children (46 children in the placebo group and 54 children in the OP0201 group). Sociodemographic and clinical characteristics of the 103 children who were randomly assigned are shown in Table 1.

Table 2 summarizes findings concerning the study’s primary and secondary endpoints. No statistically significant treatment group differences were observed for the primary endpoints (absence of bulging at day 4 [+2], no middle-ear effusion at day 12 [+2]) or for having a normal (type A) tympanogram at each of these visits. With regards to symptomatic or functional improvement, no treatment effects were apparent in the proportion of children who exhibited complete or near complete resolution of symptoms (AOM cure), defined as an AOM-SOS score ≤2, or among those who achieved ≥50% reduction from baseline at the visits on days 4 [+2], 12 [+2], and 28 [±2], or in change from baseline in AOM-SOS score over the 10-day treatment period according to parent and caregiver diary data. Finally, no clinically meaningful treatment effects were noted regarding the proportion of children categorized as treatment failure at the visit on day 12 [+2] or the proportion of children experiencing recurrences. There were no meaningful differences by treatment group in parents’ experience with using the device (data not shown).

Details concerning selected adverse events assessed as related to study treatment are summarized in Table 3; no relevant differences were noted.

We examined the safety and efficacy of intranasally administered OP0201, in addition to antimicrobial therapy, in improving outcomes (bulging of the tympanic membrane, middle-ear effusion, and resolution of symptoms) in children aged 6 to 24 months with AOM. For the trials’ primary and secondary outcomes, we found no clinically meaningful differences between treatment groups. Of interest, albeit not reaching significance, the proportion of children with middle-ear effusion was lower among children receiving OP0201 at all study visits. Overall, administration of OP0201 was safe; we found no evidence of more adverse events in OP0201 recipients compared with controls.

Our trial had certain strengths: the randomized, double-blind, placebo-controlled design, a diverse participants’ population in the age group most prone to having AOM recurrences, our use of stringent AOM diagnostic criteria, otoscopic diagnoses by validated otoscopists, a standardized antimicrobial protocol for treating episodes, validated scales for rating severity of symptoms, and modest attrition. Limitations of our trial include its small sample size and dose selection based on the maximum dose and drug concentration delivered in the metered-dose inhaler and number of sprays a parent would be willing to administer. Furthermore, of the originally estimated 140 children, 103 were enrolled in the study. As a phase 2a study, our goal was to demonstrate evidence of some clinical efficacy while monitoring for adverse events. If there is an undetected true benefit of surfactant treatment in children with AOM, the expected treatment effect is likely to be smaller than 25%. All children received adequate 10-day treatment with amoxicillin/clavulanate, potentially reducing the additional benefit an adjunct treatment such as surfactant might provide. On the basis of our previous reports on treatment of AOM in young children^2,3  we provided 10 days of antimicrobial therapy to all children. It is possible that among children not treated with antimicrobial agents, surfactant would have resulted in significant benefit compared with placebo. Although it did not appear that parents had problems with ease of administration of study medication, it was challenging for them to meet the narrow per-protocol visit windows, which were the prespecified analysis windows; accordingly, the visit days were chosen to maximize the treatment effect.

This is the first study of surfactant administered intranasally in young children with AOM. A novel therapy that potentially improves symptoms of AOM while reducing antimicrobial use or duration of therapy would be desirable. The biological basis for surfactant in the treatment of AOM derives from its ability to lower surface tension, which may open the ET usually obstructed during an AOM episode. Animal studies with surfactant have revealed significant improvements in ET opening pressures and persistence of middle-ear effusion.^14–16  In chinchillas with laboratory-induced AOM, surfactant revealed improvements in tympanometry at day 12, incidence of labyrinthitis, microbiologic cure rates, and middle-ear effusion on clinical examination.¹⁷  Three phase 1 adult studies of surfactant have been conducted^18–20 ; no pediatric studies have been reported. We were unable to reproduce the encouraging results seen in animal studies.

In young children with AOM, intranasally administered surfactant (OP0201) at the dose evaluated did not result in improved clinical outcomes. Further research, including dose ranging studies, may be warranted among children with persistent middle-ear effusion.

We are grateful to the many UPMC Children’s Hospital of Pittsburgh house officers, General Academic Pediatrics faculty and nurses who referred children to the study, and to Kris Daw and Marcia Pope who helped manage children in the study. We are particularly indebted to the children and their families for their generosity and cooperation in participating in this trial to further clinical research in children.

AOM

acute otitis media

AOM-SOS

Acute Otitis Media–Severity of Symptoms

ET

eustachian tube

mITT

modified intent-to-treat